Fix multiple OpenSSL vulnerabilities.

Security:	FreeBSD-SA-16:12.openssl
Approved by:	so

17 files changed, 1232 insertions, 469 deletions

diff --git a/UPDATING b/UPDATING
index d3b3cb92025e..0ec21dd9cd4e 100644
--- a/UPDATING
+++ b/UPDATING
@@ -16,6 +16,10 @@ from older versions of FreeBSD, try WITHOUT_CLANG to bootstrap to the tip of
 stable/10, and then rebuild without this option. The bootstrap process from
 older version of current is a bit fragile.
 
+20160303	p13	FreeBSD-SA-16:12.openssl
+
+	Fix multiple vulnerabilities of OpenSSL.
+
 20160130	p12	FreeBSD-SA-16:11.openssl
 
 	Fix OpenSSL SSLv2 ciphersuite downgrade vulnerability. [SA-16:11]
diff --git a/crypto/openssl/apps/s_server.c b/crypto/openssl/apps/s_server.c
index b58e5e07c41c..a53cadd66052 100644
--- a/crypto/openssl/apps/s_server.c
+++ b/crypto/openssl/apps/s_server.c
@@ -416,6 +416,8 @@ typedef struct srpsrvparm_st {
 static int MS_CALLBACK ssl_srp_server_param_cb(SSL *s, int *ad, void *arg)
 {
     srpsrvparm *p = (srpsrvparm *) arg;
+    int ret = SSL3_AL_FATAL;
+
     if (p->login == NULL && p->user == NULL) {
         p->login = SSL_get_srp_username(s);
         BIO_printf(bio_err, "SRP username = \"%s\"\n", p->login);
@@ -424,21 +426,25 @@ static int MS_CALLBACK ssl_srp_server_param_cb(SSL *s, int *ad, void *arg)
 
     if (p->user == NULL) {
         BIO_printf(bio_err, "User %s doesn't exist\n", p->login);
-        return SSL3_AL_FATAL;
+        goto err;
     }
+
     if (SSL_set_srp_server_param
         (s, p->user->N, p->user->g, p->user->s, p->user->v,
          p->user->info) < 0) {
         *ad = SSL_AD_INTERNAL_ERROR;
-        return SSL3_AL_FATAL;
+        goto err;
     }
     BIO_printf(bio_err,
                "SRP parameters set: username = \"%s\" info=\"%s\" \n",
                p->login, p->user->info);
-    /* need to check whether there are memory leaks */
+    ret = SSL_ERROR_NONE;
+
+err:
+    SRP_user_pwd_free(p->user);
     p->user = NULL;
     p->login = NULL;
-    return SSL_ERROR_NONE;
+    return ret;
 }
 
 #endif
@@ -2244,9 +2250,10 @@ static int sv_body(char *hostname, int s, unsigned char *context)
 #ifndef OPENSSL_NO_SRP
                 while (SSL_get_error(con, k) == SSL_ERROR_WANT_X509_LOOKUP) {
                     BIO_printf(bio_s_out, "LOOKUP renego during write\n");
+                    SRP_user_pwd_free(srp_callback_parm.user);
                     srp_callback_parm.user =
-                        SRP_VBASE_get_by_user(srp_callback_parm.vb,
-                                              srp_callback_parm.login);
+                        SRP_VBASE_get1_by_user(srp_callback_parm.vb,
+                                               srp_callback_parm.login);
                     if (srp_callback_parm.user)
                         BIO_printf(bio_s_out, "LOOKUP done %s\n",
                                    srp_callback_parm.user->info);
@@ -2300,9 +2307,10 @@ static int sv_body(char *hostname, int s, unsigned char *context)
 #ifndef OPENSSL_NO_SRP
                 while (SSL_get_error(con, i) == SSL_ERROR_WANT_X509_LOOKUP) {
                     BIO_printf(bio_s_out, "LOOKUP renego during read\n");
+                    SRP_user_pwd_free(srp_callback_parm.user);
                     srp_callback_parm.user =
-                        SRP_VBASE_get_by_user(srp_callback_parm.vb,
-                                              srp_callback_parm.login);
+                        SRP_VBASE_get1_by_user(srp_callback_parm.vb,
+                                               srp_callback_parm.login);
                     if (srp_callback_parm.user)
                         BIO_printf(bio_s_out, "LOOKUP done %s\n",
                                    srp_callback_parm.user->info);
@@ -2387,9 +2395,10 @@ static int init_ssl_connection(SSL *con)
     while (i <= 0 && SSL_get_error(con, i) == SSL_ERROR_WANT_X509_LOOKUP) {
         BIO_printf(bio_s_out, "LOOKUP during accept %s\n",
                    srp_callback_parm.login);
+        SRP_user_pwd_free(srp_callback_parm.user);
         srp_callback_parm.user =
-            SRP_VBASE_get_by_user(srp_callback_parm.vb,
-                                  srp_callback_parm.login);
+            SRP_VBASE_get1_by_user(srp_callback_parm.vb,
+                                   srp_callback_parm.login);
         if (srp_callback_parm.user)
             BIO_printf(bio_s_out, "LOOKUP done %s\n",
                        srp_callback_parm.user->info);
@@ -2616,9 +2625,10 @@ static int www_body(char *hostname, int s, unsigned char *context)
                    && SSL_get_error(con, i) == SSL_ERROR_WANT_X509_LOOKUP) {
                 BIO_printf(bio_s_out, "LOOKUP during accept %s\n",
                            srp_callback_parm.login);
+                SRP_user_pwd_free(srp_callback_parm.user);
                 srp_callback_parm.user =
-                    SRP_VBASE_get_by_user(srp_callback_parm.vb,
-                                          srp_callback_parm.login);
+                    SRP_VBASE_get1_by_user(srp_callback_parm.vb,
+                                           srp_callback_parm.login);
                 if (srp_callback_parm.user)
                     BIO_printf(bio_s_out, "LOOKUP done %s\n",
                                srp_callback_parm.user->info);
@@ -2654,6 +2664,22 @@ static int www_body(char *hostname, int s, unsigned char *context)
                 goto err;
             } else {
                 BIO_printf(bio_s_out, "read R BLOCK\n");
+#ifndef OPENSSL_NO_SRP
+                if (BIO_should_io_special(io)
+                    && BIO_get_retry_reason(io) == BIO_RR_SSL_X509_LOOKUP) {
+                    BIO_printf(bio_s_out, "LOOKUP renego during read\n");
+                    SRP_user_pwd_free(srp_callback_parm.user);
+                    srp_callback_parm.user =
+                        SRP_VBASE_get1_by_user(srp_callback_parm.vb,
+                                               srp_callback_parm.login);
+                    if (srp_callback_parm.user)
+                        BIO_printf(bio_s_out, "LOOKUP done %s\n",
+                                   srp_callback_parm.user->info);
+                    else
+                        BIO_printf(bio_s_out, "LOOKUP not successful\n");
+                    continue;
+                }
+#endif
 #if defined(OPENSSL_SYS_NETWARE)
                 delay(1000);
 #elif !defined(OPENSSL_SYS_MSDOS) && !defined(__DJGPP__)
diff --git a/crypto/openssl/crypto/bio/b_print.c b/crypto/openssl/crypto/bio/b_print.c
index 7c81e25d482c..90248fa2aaba 100644
--- a/crypto/openssl/crypto/bio/b_print.c
+++ b/crypto/openssl/crypto/bio/b_print.c
@@ -125,16 +125,16 @@
 # define LLONG long
 #endif
 
-static void fmtstr(char **, char **, size_t *, size_t *,
-                   const char *, int, int, int);
-static void fmtint(char **, char **, size_t *, size_t *,
-                   LLONG, int, int, int, int);
-static void fmtfp(char **, char **, size_t *, size_t *,
-                  LDOUBLE, int, int, int);
-static void doapr_outch(char **, char **, size_t *, size_t *, int);
-static void _dopr(char **sbuffer, char **buffer,
-                  size_t *maxlen, size_t *retlen, int *truncated,
-                  const char *format, va_list args);
+static int fmtstr(char **, char **, size_t *, size_t *,
+                  const char *, int, int, int);
+static int fmtint(char **, char **, size_t *, size_t *,
+                  LLONG, int, int, int, int);
+static int fmtfp(char **, char **, size_t *, size_t *,
+                 LDOUBLE, int, int, int);
+static int doapr_outch(char **, char **, size_t *, size_t *, int);
+static int _dopr(char **sbuffer, char **buffer,
+                 size_t *maxlen, size_t *retlen, int *truncated,
+                 const char *format, va_list args);
 
 /* format read states */
 #define DP_S_DEFAULT    0
@@ -165,7 +165,7 @@ static void _dopr(char **sbuffer, char **buffer,
 #define char_to_int(p) (p - '0')
 #define OSSL_MAX(p,q) ((p >= q) ? p : q)
 
-static void
+static int
 _dopr(char **sbuffer,
       char **buffer,
       size_t *maxlen,
@@ -196,7 +196,8 @@ _dopr(char **sbuffer,
             if (ch == '%')
                 state = DP_S_FLAGS;
             else
-                doapr_outch(sbuffer, buffer, &currlen, maxlen, ch);
+                if(!doapr_outch(sbuffer, buffer, &currlen, maxlen, ch))
+                    return 0;
             ch = *format++;
             break;
         case DP_S_FLAGS:
@@ -302,8 +303,9 @@ _dopr(char **sbuffer,
                     value = va_arg(args, int);
                     break;
                 }
-                fmtint(sbuffer, buffer, &currlen, maxlen,
-                       value, 10, min, max, flags);
+                if (!fmtint(sbuffer, buffer, &currlen, maxlen, value, 10, min,
+                            max, flags))
+                    return 0;
                 break;
             case 'X':
                 flags |= DP_F_UP;
@@ -326,17 +328,19 @@ _dopr(char **sbuffer,
                     value = (LLONG) va_arg(args, unsigned int);
                     break;
                 }
-                fmtint(sbuffer, buffer, &currlen, maxlen, value,
-                       ch == 'o' ? 8 : (ch == 'u' ? 10 : 16),
-                       min, max, flags);
+                if (!fmtint(sbuffer, buffer, &currlen, maxlen, value,
+                            ch == 'o' ? 8 : (ch == 'u' ? 10 : 16),
+                            min, max, flags))
+                    return 0;
                 break;
             case 'f':
                 if (cflags == DP_C_LDOUBLE)
                     fvalue = va_arg(args, LDOUBLE);
                 else
                     fvalue = va_arg(args, double);
-                fmtfp(sbuffer, buffer, &currlen, maxlen,
-                      fvalue, min, max, flags);
+                if (!fmtfp(sbuffer, buffer, &currlen, maxlen, fvalue, min, max,
+                           flags))
+                    return 0;
                 break;
             case 'E':
                 flags |= DP_F_UP;
@@ -355,8 +359,9 @@ _dopr(char **sbuffer,
                     fvalue = va_arg(args, double);
                 break;
             case 'c':
-                doapr_outch(sbuffer, buffer, &currlen, maxlen,
-                            va_arg(args, int));
+                if(!doapr_outch(sbuffer, buffer, &currlen, maxlen,
+                            va_arg(args, int)))
+                    return 0;
                 break;
             case 's':
                 strvalue = va_arg(args, char *);
@@ -366,13 +371,15 @@ _dopr(char **sbuffer,
                     else
                         max = *maxlen;
                 }
-                fmtstr(sbuffer, buffer, &currlen, maxlen, strvalue,
-                       flags, min, max);
+                if (!fmtstr(sbuffer, buffer, &currlen, maxlen, strvalue,
+                            flags, min, max))
+                    return 0;
                 break;
             case 'p':
                 value = (long)va_arg(args, void *);
-                fmtint(sbuffer, buffer, &currlen, maxlen,
-                       value, 16, min, max, flags | DP_F_NUM);
+                if (!fmtint(sbuffer, buffer, &currlen, maxlen,
+                            value, 16, min, max, flags | DP_F_NUM))
+                    return 0;
                 break;
             case 'n':          /* XXX */
                 if (cflags == DP_C_SHORT) {
@@ -394,7 +401,8 @@ _dopr(char **sbuffer,
                 }
                 break;
             case '%':
-                doapr_outch(sbuffer, buffer, &currlen, maxlen, ch);
+                if(!doapr_outch(sbuffer, buffer, &currlen, maxlen, ch))
+                    return 0;
                 break;
             case 'w':
                 /* not supported yet, treat as next char */
@@ -418,46 +426,56 @@ _dopr(char **sbuffer,
     *truncated = (currlen > *maxlen - 1);
     if (*truncated)
         currlen = *maxlen - 1;
-    doapr_outch(sbuffer, buffer, &currlen, maxlen, '\0');
+    if(!doapr_outch(sbuffer, buffer, &currlen, maxlen, '\0'))
+        return 0;
     *retlen = currlen - 1;
-    return;
+    return 1;
 }
 
-static void
+static int
 fmtstr(char **sbuffer,
        char **buffer,
        size_t *currlen,
        size_t *maxlen, const char *value, int flags, int min, int max)
 {
-    int padlen, strln;
+    int padlen;
+    size_t strln;
     int cnt = 0;
 
     if (value == 0)
         value = "<NULL>";
-    for (strln = 0; value[strln]; ++strln) ;
+
+    strln = strlen(value);
+    if (strln > INT_MAX)
+        strln = INT_MAX;
+
     padlen = min - strln;
-    if (padlen < 0)
+    if (min < 0 || padlen < 0)
         padlen = 0;
     if (flags & DP_F_MINUS)
         padlen = -padlen;
 
     while ((padlen > 0) && (cnt < max)) {
-        doapr_outch(sbuffer, buffer, currlen, maxlen, ' ');
+        if(!doapr_outch(sbuffer, buffer, currlen, maxlen, ' '))
+            return 0;
         --padlen;
         ++cnt;
     }
     while (*value && (cnt < max)) {
-        doapr_outch(sbuffer, buffer, currlen, maxlen, *value++);
+        if(!doapr_outch(sbuffer, buffer, currlen, maxlen, *value++))
+            return 0;
         ++cnt;
     }
     while ((padlen < 0) && (cnt < max)) {
-        doapr_outch(sbuffer, buffer, currlen, maxlen, ' ');
+        if(!doapr_outch(sbuffer, buffer, currlen, maxlen, ' '))
+            return 0;
         ++padlen;
         ++cnt;
     }
+    return 1;
 }
 
-static void
+static int
 fmtint(char **sbuffer,
        char **buffer,
        size_t *currlen,
@@ -517,37 +535,44 @@ fmtint(char **sbuffer,
 
     /* spaces */
     while (spadlen > 0) {
-        doapr_outch(sbuffer, buffer, currlen, maxlen, ' ');
+        if(!doapr_outch(sbuffer, buffer, currlen, maxlen, ' '))
+            return 0;
         --spadlen;
     }
 
     /* sign */
     if (signvalue)
-        doapr_outch(sbuffer, buffer, currlen, maxlen, signvalue);
+        if(!doapr_outch(sbuffer, buffer, currlen, maxlen, signvalue))
+            return 0;
 
     /* prefix */
     while (*prefix) {
-        doapr_outch(sbuffer, buffer, currlen, maxlen, *prefix);
+        if(!doapr_outch(sbuffer, buffer, currlen, maxlen, *prefix))
+            return 0;
         prefix++;
     }
 
     /* zeros */
     if (zpadlen > 0) {
         while (zpadlen > 0) {
-            doapr_outch(sbuffer, buffer, currlen, maxlen, '0');
+            if(!doapr_outch(sbuffer, buffer, currlen, maxlen, '0'))
+                return 0;
             --zpadlen;
         }
     }
     /* digits */
-    while (place > 0)
-        doapr_outch(sbuffer, buffer, currlen, maxlen, convert[--place]);
+    while (place > 0) {
+        if (!doapr_outch(sbuffer, buffer, currlen, maxlen, convert[--place]))
+            return 0;
+    }
 
     /* left justified spaces */
     while (spadlen < 0) {
-        doapr_outch(sbuffer, buffer, currlen, maxlen, ' ');
+        if (!doapr_outch(sbuffer, buffer, currlen, maxlen, ' '))
+            return 0;
         ++spadlen;
     }
-    return;
+    return 1;
 }
 
 static LDOUBLE abs_val(LDOUBLE value)
@@ -578,7 +603,7 @@ static long roundv(LDOUBLE value)
     return intpart;
 }
 
-static void
+static int
 fmtfp(char **sbuffer,
       char **buffer,
       size_t *currlen,
@@ -657,47 +682,61 @@ fmtfp(char **sbuffer,
 
     if ((flags & DP_F_ZERO) && (padlen > 0)) {
         if (signvalue) {
-            doapr_outch(sbuffer, buffer, currlen, maxlen, signvalue);
+            if (!doapr_outch(sbuffer, buffer, currlen, maxlen, signvalue))
+                return 0;
             --padlen;
             signvalue = 0;
         }
         while (padlen > 0) {
-            doapr_outch(sbuffer, buffer, currlen, maxlen, '0');
+            if (!doapr_outch(sbuffer, buffer, currlen, maxlen, '0'))
+                return 0;
             --padlen;
         }
     }
     while (padlen > 0) {
-        doapr_outch(sbuffer, buffer, currlen, maxlen, ' ');
+        if (!doapr_outch(sbuffer, buffer, currlen, maxlen, ' '))
+            return 0;
         --padlen;
     }
-    if (signvalue)
-        doapr_outch(sbuffer, buffer, currlen, maxlen, signvalue);
+    if (signvalue && !doapr_outch(sbuffer, buffer, currlen, maxlen, signvalue))
+        return 0;
 
-    while (iplace > 0)
-        doapr_outch(sbuffer, buffer, currlen, maxlen, iconvert[--iplace]);
+    while (iplace > 0) {
+        if (!doapr_outch(sbuffer, buffer, currlen, maxlen, iconvert[--iplace]))
+            return 0;
+    }
 
     /*
      * Decimal point. This should probably use locale to find the correct
      * char to print out.
      */
     if (max > 0 || (flags & DP_F_NUM)) {
-        doapr_outch(sbuffer, buffer, currlen, maxlen, '.');
+        if (!doapr_outch(sbuffer, buffer, currlen, maxlen, '.'))
+            return 0;
 
-        while (fplace > 0)
-            doapr_outch(sbuffer, buffer, currlen, maxlen, fconvert[--fplace]);
+        while (fplace > 0) {
+            if(!doapr_outch(sbuffer, buffer, currlen, maxlen,
+                            fconvert[--fplace]))
+                return 0;
+        }
     }
     while (zpadlen > 0) {
-        doapr_outch(sbuffer, buffer, currlen, maxlen, '0');
+        if (!doapr_outch(sbuffer, buffer, currlen, maxlen, '0'))
+            return 0;
         --zpadlen;
     }
 
     while (padlen < 0) {
-        doapr_outch(sbuffer, buffer, currlen, maxlen, ' ');
+        if (!doapr_outch(sbuffer, buffer, currlen, maxlen, ' '))
+            return 0;
         ++padlen;
     }
+    return 1;
 }
 
-static void
+#define BUFFER_INC  1024
+
+static int
 doapr_outch(char **sbuffer,
             char **buffer, size_t *currlen, size_t *maxlen, int c)
 {
@@ -708,24 +747,25 @@ doapr_outch(char **sbuffer,
     assert(*currlen <= *maxlen);
 
     if (buffer && *currlen == *maxlen) {
-        *maxlen += 1024;
+        if (*maxlen > INT_MAX - BUFFER_INC)
+            return 0;
+
+        *maxlen += BUFFER_INC;
         if (*buffer == NULL) {
             *buffer = OPENSSL_malloc(*maxlen);
-            if (!*buffer) {
-                /* Panic! Can't really do anything sensible. Just return */
-                return;
-            }
+            if (*buffer == NULL)
+                return 0;
             if (*currlen > 0) {
                 assert(*sbuffer != NULL);
                 memcpy(*buffer, *sbuffer, *currlen);
             }
             *sbuffer = NULL;
         } else {
-            *buffer = OPENSSL_realloc(*buffer, *maxlen);
-            if (!*buffer) {
-                /* Panic! Can't really do anything sensible. Just return */
-                return;
-            }
+            char *tmpbuf;
+            tmpbuf = OPENSSL_realloc(*buffer, *maxlen);
+            if (tmpbuf == NULL)
+                return 0;
+            *buffer = tmpbuf;
         }
     }
 
@@ -736,7 +776,7 @@ doapr_outch(char **sbuffer,
             (*buffer)[(*currlen)++] = (char)c;
     }
 
-    return;
+    return 1;
 }
 
 /***************************************************************************/
@@ -768,7 +808,11 @@ int BIO_vprintf(BIO *bio, const char *format, va_list args)
 
     dynbuf = NULL;
     CRYPTO_push_info("doapr()");
-    _dopr(&hugebufp, &dynbuf, &hugebufsize, &retlen, &ignored, format, args);
+    if (!_dopr(&hugebufp, &dynbuf, &hugebufsize, &retlen, &ignored, format,
+                args)) {
+        OPENSSL_free(dynbuf);
+        return -1;
+    }
     if (dynbuf) {
         ret = BIO_write(bio, dynbuf, (int)retlen);
         OPENSSL_free(dynbuf);
@@ -803,7 +847,8 @@ int BIO_vsnprintf(char *buf, size_t n, const char *format, va_list args)
     size_t retlen;
     int truncated;
 
-    _dopr(&buf, NULL, &n, &retlen, &truncated, format, args);
+    if(!_dopr(&buf, NULL, &n, &retlen, &truncated, format, args))
+        return -1;
 
     if (truncated)
         /*
diff --git a/crypto/openssl/crypto/bn/asm/x86_64-mont5.pl b/crypto/openssl/crypto/bn/asm/x86_64-mont5.pl
index dae0fe245318..235979181f54 100755
--- a/crypto/openssl/crypto/bn/asm/x86_64-mont5.pl
+++ b/crypto/openssl/crypto/bn/asm/x86_64-mont5.pl
@@ -66,60 +66,113 @@ bn_mul_mont_gather5:
 .align	16
 .Lmul_enter:
 	mov	${num}d,${num}d
-	mov	`($win64?56:8)`(%rsp),%r10d	# load 7th argument
+	movd	`($win64?56:8)`(%rsp),%xmm5	# load 7th argument
+	lea	.Linc(%rip),%r10
 	push	%rbx
 	push	%rbp
 	push	%r12
 	push	%r13
 	push	%r14
 	push	%r15
-___
-$code.=<<___ if ($win64);
-	lea	-0x28(%rsp),%rsp
-	movaps	%xmm6,(%rsp)
-	movaps	%xmm7,0x10(%rsp)
+
 .Lmul_alloca:
-___
-$code.=<<___;
 	mov	%rsp,%rax
 	lea	2($num),%r11
 	neg	%r11
-	lea	(%rsp,%r11,8),%rsp	# tp=alloca(8*(num+2))
+	lea	-264(%rsp,%r11,8),%rsp	# tp=alloca(8*(num+2)+256+8)
 	and	\$-1024,%rsp		# minimize TLB usage
 
 	mov	%rax,8(%rsp,$num,8)	# tp[num+1]=%rsp
 .Lmul_body:
-	mov	$bp,%r12		# reassign $bp
+	lea	128($bp),%r12		# reassign $bp (+size optimization)
 ___
 		$bp="%r12";
 		$STRIDE=2**5*8;		# 5 is "window size"
 		$N=$STRIDE/4;		# should match cache line size
 $code.=<<___;
-	mov	%r10,%r11
-	shr	\$`log($N/8)/log(2)`,%r10
-	and	\$`$N/8-1`,%r11
-	not	%r10
-	lea	.Lmagic_masks(%rip),%rax
-	and	\$`2**5/($N/8)-1`,%r10	# 5 is "window size"
-	lea	96($bp,%r11,8),$bp	# pointer within 1st cache line
-	movq	0(%rax,%r10,8),%xmm4	# set of masks denoting which
-	movq	8(%rax,%r10,8),%xmm5	# cache line contains element
-	movq	16(%rax,%r10,8),%xmm6	# denoted by 7th argument
-	movq	24(%rax,%r10,8),%xmm7
-
-	movq	`0*$STRIDE/4-96`($bp),%xmm0
-	movq	`1*$STRIDE/4-96`($bp),%xmm1
-	pand	%xmm4,%xmm0
-	movq	`2*$STRIDE/4-96`($bp),%xmm2
-	pand	%xmm5,%xmm1
-	movq	`3*$STRIDE/4-96`($bp),%xmm3
-	pand	%xmm6,%xmm2
-	por	%xmm1,%xmm0
-	pand	%xmm7,%xmm3
+	movdqa	0(%r10),%xmm0		# 00000001000000010000000000000000
+	movdqa	16(%r10),%xmm1		# 00000002000000020000000200000002
+	lea	24-112(%rsp,$num,8),%r10# place the mask after tp[num+3] (+ICache optimization)
+	and	\$-16,%r10
+
+	pshufd	\$0,%xmm5,%xmm5		# broadcast index
+	movdqa	%xmm1,%xmm4
+	movdqa	%xmm1,%xmm2
+___
+########################################################################
+# calculate mask by comparing 0..31 to index and save result to stack
+#
+$code.=<<___;
+	paddd	%xmm0,%xmm1
+	pcmpeqd	%xmm5,%xmm0		# compare to 1,0
+	.byte	0x67
+	movdqa	%xmm4,%xmm3
+___
+for($k=0;$k<$STRIDE/16-4;$k+=4) {
+$code.=<<___;
+	paddd	%xmm1,%xmm2
+	pcmpeqd	%xmm5,%xmm1		# compare to 3,2
+	movdqa	%xmm0,`16*($k+0)+112`(%r10)
+	movdqa	%xmm4,%xmm0
+
+	paddd	%xmm2,%xmm3
+	pcmpeqd	%xmm5,%xmm2		# compare to 5,4
+	movdqa	%xmm1,`16*($k+1)+112`(%r10)
+	movdqa	%xmm4,%xmm1
+
+	paddd	%xmm3,%xmm0
+	pcmpeqd	%xmm5,%xmm3		# compare to 7,6
+	movdqa	%xmm2,`16*($k+2)+112`(%r10)
+	movdqa	%xmm4,%xmm2
+
+	paddd	%xmm0,%xmm1
+	pcmpeqd	%xmm5,%xmm0
+	movdqa	%xmm3,`16*($k+3)+112`(%r10)
+	movdqa	%xmm4,%xmm3
+___
+}
+$code.=<<___;				# last iteration can be optimized
+	paddd	%xmm1,%xmm2
+	pcmpeqd	%xmm5,%xmm1
+	movdqa	%xmm0,`16*($k+0)+112`(%r10)
+
+	paddd	%xmm2,%xmm3
+	.byte	0x67
+	pcmpeqd	%xmm5,%xmm2
+	movdqa	%xmm1,`16*($k+1)+112`(%r10)
+
+	pcmpeqd	%xmm5,%xmm3
+	movdqa	%xmm2,`16*($k+2)+112`(%r10)
+	pand	`16*($k+0)-128`($bp),%xmm0	# while it's still in register
+
+	pand	`16*($k+1)-128`($bp),%xmm1
+	pand	`16*($k+2)-128`($bp),%xmm2
+	movdqa	%xmm3,`16*($k+3)+112`(%r10)
+	pand	`16*($k+3)-128`($bp),%xmm3
+	por	%xmm2,%xmm0
+	por	%xmm3,%xmm1
+___
+for($k=0;$k<$STRIDE/16-4;$k+=4) {
+$code.=<<___;
+	movdqa	`16*($k+0)-128`($bp),%xmm4
+	movdqa	`16*($k+1)-128`($bp),%xmm5
+	movdqa	`16*($k+2)-128`($bp),%xmm2
+	pand	`16*($k+0)+112`(%r10),%xmm4
+	movdqa	`16*($k+3)-128`($bp),%xmm3
+	pand	`16*($k+1)+112`(%r10),%xmm5
+	por	%xmm4,%xmm0
+	pand	`16*($k+2)+112`(%r10),%xmm2
+	por	%xmm5,%xmm1
+	pand	`16*($k+3)+112`(%r10),%xmm3
 	por	%xmm2,%xmm0
+	por	%xmm3,%xmm1
+___
+}
+$code.=<<___;
+	por	%xmm1,%xmm0
+	pshufd	\$0x4e,%xmm0,%xmm1
+	por	%xmm1,%xmm0
 	lea	$STRIDE($bp),$bp
-	por	%xmm3,%xmm0
-
 	movq	%xmm0,$m0		# m0=bp[0]
 
 	mov	($n0),$n0		# pull n0[0] value
@@ -128,29 +181,14 @@ $code.=<<___;
 	xor	$i,$i			# i=0
 	xor	$j,$j			# j=0
 
-	movq	`0*$STRIDE/4-96`($bp),%xmm0
-	movq	`1*$STRIDE/4-96`($bp),%xmm1
-	pand	%xmm4,%xmm0
-	movq	`2*$STRIDE/4-96`($bp),%xmm2
-	pand	%xmm5,%xmm1
-
 	mov	$n0,$m1
 	mulq	$m0			# ap[0]*bp[0]
 	mov	%rax,$lo0
 	mov	($np),%rax
 
-	movq	`3*$STRIDE/4-96`($bp),%xmm3
-	pand	%xmm6,%xmm2
-	por	%xmm1,%xmm0
-	pand	%xmm7,%xmm3
-
 	imulq	$lo0,$m1		# "tp[0]"*n0
 	mov	%rdx,$hi0
 
-	por	%xmm2,%xmm0
-	lea	$STRIDE($bp),$bp
-	por	%xmm3,%xmm0
-
 	mulq	$m1			# np[0]*m1
 	add	%rax,$lo0		# discarded
 	mov	8($ap),%rax
@@ -183,8 +221,6 @@ $code.=<<___;
 	cmp	$num,$j
 	jne	.L1st
 
-	movq	%xmm0,$m0		# bp[1]
-
 	add	%rax,$hi1
 	mov	($ap),%rax		# ap[0]
 	adc	\$0,%rdx
@@ -204,33 +240,46 @@ $code.=<<___;
 	jmp	.Louter
 .align	16
 .Louter:
+	lea	24+128(%rsp,$num,8),%rdx	# where 256-byte mask is (+size optimization)
+	and	\$-16,%rdx
+	pxor	%xmm4,%xmm4
+	pxor	%xmm5,%xmm5
+___
+for($k=0;$k<$STRIDE/16;$k+=4) {
+$code.=<<___;
+	movdqa	`16*($k+0)-128`($bp),%xmm0
+	movdqa	`16*($k+1)-128`($bp),%xmm1
+	movdqa	`16*($k+2)-128`($bp),%xmm2
+	movdqa	`16*($k+3)-128`($bp),%xmm3
+	pand	`16*($k+0)-128`(%rdx),%xmm0
+	pand	`16*($k+1)-128`(%rdx),%xmm1
+	por	%xmm0,%xmm4
+	pand	`16*($k+2)-128`(%rdx),%xmm2
+	por	%xmm1,%xmm5
+	pand	`16*($k+3)-128`(%rdx),%xmm3
+	por	%xmm2,%xmm4
+	por	%xmm3,%xmm5
+___
+}
+$code.=<<___;
+	por	%xmm5,%xmm4
+	pshufd	\$0x4e,%xmm4,%xmm0
+	por	%xmm4,%xmm0
+	lea	$STRIDE($bp),$bp
+	movq	%xmm0,$m0		# m0=bp[i]
+
 	xor	$j,$j			# j=0
 	mov	$n0,$m1
 	mov	(%rsp),$lo0
 
-	movq	`0*$STRIDE/4-96`($bp),%xmm0
-	movq	`1*$STRIDE/4-96`($bp),%xmm1
-	pand	%xmm4,%xmm0
-	movq	`2*$STRIDE/4-96`($bp),%xmm2
-	pand	%xmm5,%xmm1
-
 	mulq	$m0			# ap[0]*bp[i]
 	add	%rax,$lo0		# ap[0]*bp[i]+tp[0]
 	mov	($np),%rax
 	adc	\$0,%rdx
 
-	movq	`3*$STRIDE/4-96`($bp),%xmm3
-	pand	%xmm6,%xmm2
-	por	%xmm1,%xmm0
-	pand	%xmm7,%xmm3
-
 	imulq	$lo0,$m1		# tp[0]*n0
 	mov	%rdx,$hi0
 
-	por	%xmm2,%xmm0
-	lea	$STRIDE($bp),$bp
-	por	%xmm3,%xmm0
-
 	mulq	$m1			# np[0]*m1
 	add	%rax,$lo0		# discarded
 	mov	8($ap),%rax
@@ -266,8 +315,6 @@ $code.=<<___;
 	cmp	$num,$j
 	jne	.Linner
 
-	movq	%xmm0,$m0		# bp[i+1]
-
 	add	%rax,$hi1
 	mov	($ap),%rax		# ap[0]
 	adc	\$0,%rdx
@@ -321,13 +368,7 @@ $code.=<<___;
 
 	mov	8(%rsp,$num,8),%rsi	# restore %rsp
 	mov	\$1,%rax
-___
-$code.=<<___ if ($win64);
-	movaps	(%rsi),%xmm6
-	movaps	0x10(%rsi),%xmm7
-	lea	0x28(%rsi),%rsi
-___
-$code.=<<___;
+
 	mov	(%rsi),%r15
 	mov	8(%rsi),%r14
 	mov	16(%rsi),%r13
@@ -348,91 +389,130 @@ $code.=<<___;
 bn_mul4x_mont_gather5:
 .Lmul4x_enter:
 	mov	${num}d,${num}d
-	mov	`($win64?56:8)`(%rsp),%r10d	# load 7th argument
+	movd	`($win64?56:8)`(%rsp),%xmm5	# load 7th argument
+	lea	.Linc(%rip),%r10
 	push	%rbx
 	push	%rbp
 	push	%r12
 	push	%r13
 	push	%r14
 	push	%r15
-___
-$code.=<<___ if ($win64);
-	lea	-0x28(%rsp),%rsp
-	movaps	%xmm6,(%rsp)
-	movaps	%xmm7,0x10(%rsp)
+
 .Lmul4x_alloca:
-___
-$code.=<<___;
 	mov	%rsp,%rax
 	lea	4($num),%r11
 	neg	%r11
-	lea	(%rsp,%r11,8),%rsp	# tp=alloca(8*(num+4))
+	lea	-256(%rsp,%r11,8),%rsp	# tp=alloca(8*(num+4)+256)
 	and	\$-1024,%rsp		# minimize TLB usage
 
 	mov	%rax,8(%rsp,$num,8)	# tp[num+1]=%rsp
 .Lmul4x_body:
 	mov	$rp,16(%rsp,$num,8)	# tp[num+2]=$rp
-	mov	%rdx,%r12		# reassign $bp
+	lea	128(%rdx),%r12		# reassign $bp (+size optimization)
 ___
 		$bp="%r12";
 		$STRIDE=2**5*8;		# 5 is "window size"
 		$N=$STRIDE/4;		# should match cache line size
 $code.=<<___;
-	mov	%r10,%r11
-	shr	\$`log($N/8)/log(2)`,%r10
-	and	\$`$N/8-1`,%r11
-	not	%r10
-	lea	.Lmagic_masks(%rip),%rax
-	and	\$`2**5/($N/8)-1`,%r10	# 5 is "window size"
-	lea	96($bp,%r11,8),$bp	# pointer within 1st cache line
-	movq	0(%rax,%r10,8),%xmm4	# set of masks denoting which
-	movq	8(%rax,%r10,8),%xmm5	# cache line contains element
-	movq	16(%rax,%r10,8),%xmm6	# denoted by 7th argument
-	movq	24(%rax,%r10,8),%xmm7
-
-	movq	`0*$STRIDE/4-96`($bp),%xmm0
-	movq	`1*$STRIDE/4-96`($bp),%xmm1
-	pand	%xmm4,%xmm0
-	movq	`2*$STRIDE/4-96`($bp),%xmm2
-	pand	%xmm5,%xmm1
-	movq	`3*$STRIDE/4-96`($bp),%xmm3
-	pand	%xmm6,%xmm2
-	por	%xmm1,%xmm0
-	pand	%xmm7,%xmm3
+	movdqa	0(%r10),%xmm0		# 00000001000000010000000000000000
+	movdqa	16(%r10),%xmm1		# 00000002000000020000000200000002
+	lea	32-112(%rsp,$num,8),%r10# place the mask after tp[num+4] (+ICache optimization)
+
+	pshufd	\$0,%xmm5,%xmm5		# broadcast index
+	movdqa	%xmm1,%xmm4
+	.byte	0x67,0x67
+	movdqa	%xmm1,%xmm2
+___
+########################################################################
+# calculate mask by comparing 0..31 to index and save result to stack
+#
+$code.=<<___;
+	paddd	%xmm0,%xmm1
+	pcmpeqd	%xmm5,%xmm0		# compare to 1,0
+	.byte	0x67
+	movdqa	%xmm4,%xmm3
+___
+for($k=0;$k<$STRIDE/16-4;$k+=4) {
+$code.=<<___;
+	paddd	%xmm1,%xmm2
+	pcmpeqd	%xmm5,%xmm1		# compare to 3,2
+	movdqa	%xmm0,`16*($k+0)+112`(%r10)
+	movdqa	%xmm4,%xmm0
+
+	paddd	%xmm2,%xmm3
+	pcmpeqd	%xmm5,%xmm2		# compare to 5,4
+	movdqa	%xmm1,`16*($k+1)+112`(%r10)
+	movdqa	%xmm4,%xmm1
+
+	paddd	%xmm3,%xmm0
+	pcmpeqd	%xmm5,%xmm3		# compare to 7,6
+	movdqa	%xmm2,`16*($k+2)+112`(%r10)
+	movdqa	%xmm4,%xmm2
+
+	paddd	%xmm0,%xmm1
+	pcmpeqd	%xmm5,%xmm0
+	movdqa	%xmm3,`16*($k+3)+112`(%r10)
+	movdqa	%xmm4,%xmm3
+___
+}
+$code.=<<___;				# last iteration can be optimized
+	paddd	%xmm1,%xmm2
+	pcmpeqd	%xmm5,%xmm1
+	movdqa	%xmm0,`16*($k+0)+112`(%r10)
+
+	paddd	%xmm2,%xmm3
+	.byte	0x67
+	pcmpeqd	%xmm5,%xmm2
+	movdqa	%xmm1,`16*($k+1)+112`(%r10)
+
+	pcmpeqd	%xmm5,%xmm3
+	movdqa	%xmm2,`16*($k+2)+112`(%r10)
+	pand	`16*($k+0)-128`($bp),%xmm0	# while it's still in register
+
+	pand	`16*($k+1)-128`($bp),%xmm1
+	pand	`16*($k+2)-128`($bp),%xmm2
+	movdqa	%xmm3,`16*($k+3)+112`(%r10)
+	pand	`16*($k+3)-128`($bp),%xmm3
+	por	%xmm2,%xmm0
+	por	%xmm3,%xmm1
+___
+for($k=0;$k<$STRIDE/16-4;$k+=4) {
+$code.=<<___;
+	movdqa	`16*($k+0)-128`($bp),%xmm4
+	movdqa	`16*($k+1)-128`($bp),%xmm5
+	movdqa	`16*($k+2)-128`($bp),%xmm2
+	pand	`16*($k+0)+112`(%r10),%xmm4
+	movdqa	`16*($k+3)-128`($bp),%xmm3
+	pand	`16*($k+1)+112`(%r10),%xmm5
+	por	%xmm4,%xmm0
+	pand	`16*($k+2)+112`(%r10),%xmm2
+	por	%xmm5,%xmm1
+	pand	`16*($k+3)+112`(%r10),%xmm3
 	por	%xmm2,%xmm0
+	por	%xmm3,%xmm1
+___
+}
+$code.=<<___;
+	por	%xmm1,%xmm0
+	pshufd	\$0x4e,%xmm0,%xmm1
+	por	%xmm1,%xmm0
 	lea	$STRIDE($bp),$bp
-	por	%xmm3,%xmm0
-
 	movq	%xmm0,$m0		# m0=bp[0]
+
 	mov	($n0),$n0		# pull n0[0] value
 	mov	($ap),%rax
 
 	xor	$i,$i			# i=0
 	xor	$j,$j			# j=0
 
-	movq	`0*$STRIDE/4-96`($bp),%xmm0
-	movq	`1*$STRIDE/4-96`($bp),%xmm1
-	pand	%xmm4,%xmm0
-	movq	`2*$STRIDE/4-96`($bp),%xmm2
-	pand	%xmm5,%xmm1
-
 	mov	$n0,$m1
 	mulq	$m0			# ap[0]*bp[0]
 	mov	%rax,$A[0]
 	mov	($np),%rax
 
-	movq	`3*$STRIDE/4-96`($bp),%xmm3
-	pand	%xmm6,%xmm2
-	por	%xmm1,%xmm0
-	pand	%xmm7,%xmm3
-
 	imulq	$A[0],$m1		# "tp[0]"*n0
 	mov	%rdx,$A[1]
 
-	por	%xmm2,%xmm0
-	lea	$STRIDE($bp),$bp
-	por	%xmm3,%xmm0
-
 	mulq	$m1			# np[0]*m1
 	add	%rax,$A[0]		# discarded
 	mov	8($ap),%rax
@@ -550,8 +630,6 @@ $code.=<<___;
 	mov	$N[1],-16(%rsp,$j,8)	# tp[j-1]
 	mov	%rdx,$N[0]
 
-	movq	%xmm0,$m0		# bp[1]
-
 	xor	$N[1],$N[1]
 	add	$A[0],$N[0]
 	adc	\$0,$N[1]
@@ -561,12 +639,34 @@ $code.=<<___;
 	lea	1($i),$i		# i++
 .align	4
 .Louter4x:
+	lea	32+128(%rsp,$num,8),%rdx	# where 256-byte mask is (+size optimization)
+	pxor	%xmm4,%xmm4
+	pxor	%xmm5,%xmm5
+___
+for($k=0;$k<$STRIDE/16;$k+=4) {
+$code.=<<___;
+	movdqa	`16*($k+0)-128`($bp),%xmm0
+	movdqa	`16*($k+1)-128`($bp),%xmm1
+	movdqa	`16*($k+2)-128`($bp),%xmm2
+	movdqa	`16*($k+3)-128`($bp),%xmm3
+	pand	`16*($k+0)-128`(%rdx),%xmm0
+	pand	`16*($k+1)-128`(%rdx),%xmm1
+	por	%xmm0,%xmm4
+	pand	`16*($k+2)-128`(%rdx),%xmm2
+	por	%xmm1,%xmm5
+	pand	`16*($k+3)-128`(%rdx),%xmm3
+	por	%xmm2,%xmm4
+	por	%xmm3,%xmm5
+___
+}
+$code.=<<___;
+	por	%xmm5,%xmm4
+	pshufd	\$0x4e,%xmm4,%xmm0
+	por	%xmm4,%xmm0
+	lea	$STRIDE($bp),$bp
+	movq	%xmm0,$m0		# m0=bp[i]
+
 	xor	$j,$j			# j=0
-	movq	`0*$STRIDE/4-96`($bp),%xmm0
-	movq	`1*$STRIDE/4-96`($bp),%xmm1
-	pand	%xmm4,%xmm0
-	movq	`2*$STRIDE/4-96`($bp),%xmm2
-	pand	%xmm5,%xmm1
 
 	mov	(%rsp),$A[0]
 	mov	$n0,$m1
@@ -575,18 +675,9 @@ $code.=<<___;
 	mov	($np),%rax
 	adc	\$0,%rdx
 
-	movq	`3*$STRIDE/4-96`($bp),%xmm3
-	pand	%xmm6,%xmm2
-	por	%xmm1,%xmm0
-	pand	%xmm7,%xmm3
-
 	imulq	$A[0],$m1		# tp[0]*n0
 	mov	%rdx,$A[1]
 
-	por	%xmm2,%xmm0
-	lea	$STRIDE($bp),$bp
-	por	%xmm3,%xmm0
-
 	mulq	$m1			# np[0]*m1
 	add	%rax,$A[0]		# "$N[0]", discarded
 	mov	8($ap),%rax
@@ -718,7 +809,6 @@ $code.=<<___;
 	mov	$N[0],-24(%rsp,$j,8)	# tp[j-1]
 	mov	%rdx,$N[0]
 
-	movq	%xmm0,$m0		# bp[i+1]
 	mov	$N[1],-16(%rsp,$j,8)	# tp[j-1]
 
 	xor	$N[1],$N[1]
@@ -809,13 +899,7 @@ ___
 $code.=<<___;
 	mov	8(%rsp,$num,8),%rsi	# restore %rsp
 	mov	\$1,%rax
-___
-$code.=<<___ if ($win64);
-	movaps	(%rsi),%xmm6
-	movaps	0x10(%rsi),%xmm7
-	lea	0x28(%rsi),%rsi
-___
-$code.=<<___;
+
 	mov	(%rsi),%r15
 	mov	8(%rsi),%r14
 	mov	16(%rsi),%r13
@@ -830,8 +914,8 @@ ___
 }}}
 
 {
-my ($inp,$num,$tbl,$idx)=$win64?("%rcx","%rdx","%r8", "%r9") : # Win64 order
-				("%rdi","%rsi","%rdx","%rcx"); # Unix order
+my ($inp,$num,$tbl,$idx)=$win64?("%rcx","%rdx","%r8", "%r9d") : # Win64 order
+				("%rdi","%rsi","%rdx","%ecx"); # Unix order
 my $out=$inp;
 my $STRIDE=2**5*8;
 my $N=$STRIDE/4;
@@ -859,53 +943,89 @@ bn_scatter5:
 .type	bn_gather5,\@abi-omnipotent
 .align	16
 bn_gather5:
-___
-$code.=<<___ if ($win64);
-.LSEH_begin_bn_gather5:
+.LSEH_begin_bn_gather5:			# Win64 thing, but harmless in other cases
 	# I can't trust assembler to use specific encoding:-(
-	.byte	0x48,0x83,0xec,0x28		#sub	\$0x28,%rsp
-	.byte	0x0f,0x29,0x34,0x24		#movaps	%xmm6,(%rsp)
-	.byte	0x0f,0x29,0x7c,0x24,0x10	#movdqa	%xmm7,0x10(%rsp)
+	.byte	0x4c,0x8d,0x14,0x24			# lea    (%rsp),%r10
+	.byte	0x48,0x81,0xec,0x08,0x01,0x00,0x00	# sub	$0x108,%rsp
+	lea	.Linc(%rip),%rax
+	and	\$-16,%rsp		# shouldn't be formally required
+
+	movd	$idx,%xmm5
+	movdqa	0(%rax),%xmm0		# 00000001000000010000000000000000
+	movdqa	16(%rax),%xmm1		# 00000002000000020000000200000002
+	lea	128($tbl),%r11		# size optimization
+	lea	128(%rsp),%rax		# size optimization
+
+	pshufd	\$0,%xmm5,%xmm5		# broadcast $idx
+	movdqa	%xmm1,%xmm4
+	movdqa	%xmm1,%xmm2
 ___
+########################################################################
+# calculate mask by comparing 0..31 to $idx and save result to stack
+#
+for($i=0;$i<$STRIDE/16;$i+=4) {
+$code.=<<___;
+	paddd	%xmm0,%xmm1
+	pcmpeqd	%xmm5,%xmm0		# compare to 1,0
+___
+$code.=<<___	if ($i);
+	movdqa	%xmm3,`16*($i-1)-128`(%rax)
+___
+$code.=<<___;
+	movdqa	%xmm4,%xmm3
+
+	paddd	%xmm1,%xmm2
+	pcmpeqd	%xmm5,%xmm1		# compare to 3,2
+	movdqa	%xmm0,`16*($i+0)-128`(%rax)
+	movdqa	%xmm4,%xmm0
+
+	paddd	%xmm2,%xmm3
+	pcmpeqd	%xmm5,%xmm2		# compare to 5,4
+	movdqa	%xmm1,`16*($i+1)-128`(%rax)
+	movdqa	%xmm4,%xmm1
+
+	paddd	%xmm3,%xmm0
+	pcmpeqd	%xmm5,%xmm3		# compare to 7,6
+	movdqa	%xmm2,`16*($i+2)-128`(%rax)
+	movdqa	%xmm4,%xmm2
+___
+}
 $code.=<<___;
-	mov	$idx,%r11
-	shr	\$`log($N/8)/log(2)`,$idx
-	and	\$`$N/8-1`,%r11
-	not	$idx
-	lea	.Lmagic_masks(%rip),%rax
-	and	\$`2**5/($N/8)-1`,$idx	# 5 is "window size"
-	lea	96($tbl,%r11,8),$tbl	# pointer within 1st cache line
-	movq	0(%rax,$idx,8),%xmm4	# set of masks denoting which
-	movq	8(%rax,$idx,8),%xmm5	# cache line contains element
-	movq	16(%rax,$idx,8),%xmm6	# denoted by 7th argument
-	movq	24(%rax,$idx,8),%xmm7
+	movdqa	%xmm3,`16*($i-1)-128`(%rax)
 	jmp	.Lgather
-.align	16
-.Lgather:
-	movq	`0*$STRIDE/4-96`($tbl),%xmm0
-	movq	`1*$STRIDE/4-96`($tbl),%xmm1
-	pand	%xmm4,%xmm0
-	movq	`2*$STRIDE/4-96`($tbl),%xmm2
-	pand	%xmm5,%xmm1
-	movq	`3*$STRIDE/4-96`($tbl),%xmm3
-	pand	%xmm6,%xmm2
-	por	%xmm1,%xmm0
-	pand	%xmm7,%xmm3
-	por	%xmm2,%xmm0
-	lea	$STRIDE($tbl),$tbl
-	por	%xmm3,%xmm0
 
+.align	32
+.Lgather:
+	pxor	%xmm4,%xmm4
+	pxor	%xmm5,%xmm5
+___
+for($i=0;$i<$STRIDE/16;$i+=4) {
+$code.=<<___;
+	movdqa	`16*($i+0)-128`(%r11),%xmm0
+	movdqa	`16*($i+1)-128`(%r11),%xmm1
+	movdqa	`16*($i+2)-128`(%r11),%xmm2
+	pand	`16*($i+0)-128`(%rax),%xmm0
+	movdqa	`16*($i+3)-128`(%r11),%xmm3
+	pand	`16*($i+1)-128`(%rax),%xmm1
+	por	%xmm0,%xmm4
+	pand	`16*($i+2)-128`(%rax),%xmm2
+	por	%xmm1,%xmm5
+	pand	`16*($i+3)-128`(%rax),%xmm3
+	por	%xmm2,%xmm4
+	por	%xmm3,%xmm5
+___
+}
+$code.=<<___;
+	por	%xmm5,%xmm4
+	lea	$STRIDE(%r11),%r11
+	pshufd	\$0x4e,%xmm4,%xmm0
+	por	%xmm4,%xmm0
 	movq	%xmm0,($out)		# m0=bp[0]
 	lea	8($out),$out
 	sub	\$1,$num
 	jnz	.Lgather
-___
-$code.=<<___ if ($win64);
-	movaps	(%rsp),%xmm6
-	movaps	0x10(%rsp),%xmm7
-	lea	0x28(%rsp),%rsp
-___
-$code.=<<___;
+
+	lea	(%r10),%rsp
 	ret
 .LSEH_end_bn_gather5:
 .size	bn_gather5,.-bn_gather5
@@ -913,9 +1033,9 @@ ___
 }
 $code.=<<___;
 .align	64
-.Lmagic_masks:
-	.long	0,0, 0,0, 0,0, -1,-1
-	.long	0,0, 0,0, 0,0,  0,0
+.Linc:
+	.long	0,0, 1,1
+	.long	2,2, 2,2
 .asciz	"Montgomery Multiplication with scatter/gather for x86_64, CRYPTOGAMS by <appro\@openssl.org>"
 ___
 
@@ -954,7 +1074,7 @@ mul_handler:
 	cmp	%r10,%rbx		# context->Rip<end of prologue label
 	jb	.Lcommon_seh_tail
 
-	lea	`40+48`(%rax),%rax
+	lea	48(%rax),%rax
 
 	mov	4(%r11),%r10d		# HandlerData[1]
 	lea	(%rsi,%r10),%r10	# end of alloca label
@@ -971,9 +1091,7 @@ mul_handler:
 	mov	192($context),%r10	# pull $num
 	mov	8(%rax,%r10,8),%rax	# pull saved stack pointer
 
-	movaps	(%rax),%xmm0
-	movaps	16(%rax),%xmm1
-	lea	`40+48`(%rax),%rax
+	lea	48(%rax),%rax
 
 	mov	-8(%rax),%rbx
 	mov	-16(%rax),%rbp
@@ -987,8 +1105,6 @@ mul_handler:
 	mov	%r13,224($context)	# restore context->R13
 	mov	%r14,232($context)	# restore context->R14
 	mov	%r15,240($context)	# restore context->R15
-	movups	%xmm0,512($context)	# restore context->Xmm6
-	movups	%xmm1,528($context)	# restore context->Xmm7
 
 .Lcommon_seh_tail:
 	mov	8(%rax),%rdi
@@ -1057,10 +1173,9 @@ mul_handler:
 	.rva	.Lmul4x_alloca,.Lmul4x_body,.Lmul4x_epilogue	# HandlerData[]
 .align	8
 .LSEH_info_bn_gather5:
-        .byte   0x01,0x0d,0x05,0x00
-        .byte   0x0d,0x78,0x01,0x00	#movaps	0x10(rsp),xmm7
-        .byte   0x08,0x68,0x00,0x00	#movaps	(rsp),xmm6
-        .byte   0x04,0x42,0x00,0x00	#sub	rsp,0x28
+	.byte	0x01,0x0b,0x03,0x0a
+	.byte	0x0b,0x01,0x21,0x00	# sub	rsp,0x108
+	.byte	0x04,0xa3,0x00,0x00	# lea	r10,(rsp), set_frame r10
 .align	8
 ___
 }
diff --git a/crypto/openssl/crypto/bn/bn.h b/crypto/openssl/crypto/bn/bn.h
index 47d8c71d9ed1..b39258d4d4ac 100644
--- a/crypto/openssl/crypto/bn/bn.h
+++ b/crypto/openssl/crypto/bn/bn.h
@@ -125,6 +125,7 @@
 #ifndef HEADER_BN_H
 # define HEADER_BN_H
 
+# include <limits.h>
 # include <openssl/e_os2.h>
 # ifndef OPENSSL_NO_FP_API
 #  include <stdio.h>            /* FILE */
@@ -739,8 +740,17 @@ const BIGNUM *BN_get0_nist_prime_521(void);
 
 /* library internal functions */
 
-# define bn_expand(a,bits) ((((((bits+BN_BITS2-1))/BN_BITS2)) <= (a)->dmax)?\
-        (a):bn_expand2((a),(bits+BN_BITS2-1)/BN_BITS2))
+# define bn_expand(a,bits) \
+    ( \
+        bits > (INT_MAX - BN_BITS2 + 1) ? \
+            NULL \
+        : \
+            (((bits+BN_BITS2-1)/BN_BITS2) <= (a)->dmax) ? \
+                (a) \
+            : \
+                bn_expand2((a),(bits+BN_BITS2-1)/BN_BITS2) \
+    )
+
 # define bn_wexpand(a,words) (((words) <= (a)->dmax)?(a):bn_expand2((a),(words)))
 BIGNUM *bn_expand2(BIGNUM *a, int words);
 # ifndef OPENSSL_NO_DEPRECATED
diff --git a/crypto/openssl/crypto/bn/bn_exp.c b/crypto/openssl/crypto/bn/bn_exp.c
index 27146c89e740..52ab0b23f756 100644
--- a/crypto/openssl/crypto/bn/bn_exp.c
+++ b/crypto/openssl/crypto/bn/bn_exp.c
@@ -110,6 +110,7 @@
  */
 
 #include "cryptlib.h"
+#include "constant_time_locl.h"
 #include "bn_lcl.h"
 
 #include <stdlib.h>
@@ -535,15 +536,17 @@ int BN_mod_exp_mont(BIGNUM *rr, const BIGNUM *a, const BIGNUM *p,
 
 static int MOD_EXP_CTIME_COPY_TO_PREBUF(const BIGNUM *b, int top,
                                         unsigned char *buf, int idx,
-                                        int width)
+                                        int window)
 {
-    size_t i, j;
+    int i, j;
+    int width = 1 << window;
+    BN_ULONG *table = (BN_ULONG *)buf;
 
     if (top > b->top)
         top = b->top;           /* this works because 'buf' is explicitly
                                  * zeroed */
-    for (i = 0, j = idx; i < top * sizeof b->d[0]; i++, j += width) {
-        buf[j] = ((unsigned char *)b->d)[i];
+    for (i = 0, j = idx; i < top; i++, j += width) {
+        table[j] = b->d[i];
     }
 
     return 1;
@@ -551,15 +554,51 @@ static int MOD_EXP_CTIME_COPY_TO_PREBUF(const BIGNUM *b, int top,
 
 static int MOD_EXP_CTIME_COPY_FROM_PREBUF(BIGNUM *b, int top,
                                           unsigned char *buf, int idx,
-                                          int width)
+                                          int window)
 {
-    size_t i, j;
+    int i, j;
+    int width = 1 << window;
+    volatile BN_ULONG *table = (volatile BN_ULONG *)buf;
 
     if (bn_wexpand(b, top) == NULL)
         return 0;
 
-    for (i = 0, j = idx; i < top * sizeof b->d[0]; i++, j += width) {
-        ((unsigned char *)b->d)[i] = buf[j];
+    if (window <= 3) {
+        for (i = 0; i < top; i++, table += width) {
+            BN_ULONG acc = 0;
+
+            for (j = 0; j < width; j++) {
+                acc |= table[j] &
+                       ((BN_ULONG)0 - (constant_time_eq_int(j,idx)&1));
+            }
+
+            b->d[i] = acc;
+        }
+    } else {
+        int xstride = 1 << (window - 2);
+        BN_ULONG y0, y1, y2, y3;
+
+        i = idx >> (window - 2);        /* equivalent of idx / xstride */
+        idx &= xstride - 1;             /* equivalent of idx % xstride */
+
+        y0 = (BN_ULONG)0 - (constant_time_eq_int(i,0)&1);
+        y1 = (BN_ULONG)0 - (constant_time_eq_int(i,1)&1);
+        y2 = (BN_ULONG)0 - (constant_time_eq_int(i,2)&1);
+        y3 = (BN_ULONG)0 - (constant_time_eq_int(i,3)&1);
+
+        for (i = 0; i < top; i++, table += width) {
+            BN_ULONG acc = 0;
+
+            for (j = 0; j < xstride; j++) {
+                acc |= ( (table[j + 0 * xstride] & y0) |
+                         (table[j + 1 * xstride] & y1) |
+                         (table[j + 2 * xstride] & y2) |
+                         (table[j + 3 * xstride] & y3) )
+                       & ((BN_ULONG)0 - (constant_time_eq_int(j,idx)&1));
+            }
+
+            b->d[i] = acc;
+        }
     }
 
     b->top = top;
@@ -782,9 +821,9 @@ int BN_mod_exp_mont_consttime(BIGNUM *rr, const BIGNUM *a, const BIGNUM *p,
     } else
 #endif
     {
-        if (!MOD_EXP_CTIME_COPY_TO_PREBUF(&tmp, top, powerbuf, 0, numPowers))
+        if (!MOD_EXP_CTIME_COPY_TO_PREBUF(&tmp, top, powerbuf, 0, window))
             goto err;
-        if (!MOD_EXP_CTIME_COPY_TO_PREBUF(&am, top, powerbuf, 1, numPowers))
+        if (!MOD_EXP_CTIME_COPY_TO_PREBUF(&am, top, powerbuf, 1, window))
             goto err;
 
         /*
@@ -796,15 +835,15 @@ int BN_mod_exp_mont_consttime(BIGNUM *rr, const BIGNUM *a, const BIGNUM *p,
         if (window > 1) {
             if (!BN_mod_mul_montgomery(&tmp, &am, &am, mont, ctx))
                 goto err;
-            if (!MOD_EXP_CTIME_COPY_TO_PREBUF
-                (&tmp, top, powerbuf, 2, numPowers))
+            if (!MOD_EXP_CTIME_COPY_TO_PREBUF(&tmp, top, powerbuf, 2,
+                                              window))
                 goto err;
             for (i = 3; i < numPowers; i++) {
                 /* Calculate a^i = a^(i-1) * a */
                 if (!BN_mod_mul_montgomery(&tmp, &am, &tmp, mont, ctx))
                     goto err;
-                if (!MOD_EXP_CTIME_COPY_TO_PREBUF
-                    (&tmp, top, powerbuf, i, numPowers))
+                if (!MOD_EXP_CTIME_COPY_TO_PREBUF(&tmp, top, powerbuf, i,
+                                                  window))
                     goto err;
             }
         }
@@ -812,8 +851,8 @@ int BN_mod_exp_mont_consttime(BIGNUM *rr, const BIGNUM *a, const BIGNUM *p,
         bits--;
         for (wvalue = 0, i = bits % window; i >= 0; i--, bits--)
             wvalue = (wvalue << 1) + BN_is_bit_set(p, bits);
-        if (!MOD_EXP_CTIME_COPY_FROM_PREBUF
-            (&tmp, top, powerbuf, wvalue, numPowers))
+        if (!MOD_EXP_CTIME_COPY_FROM_PREBUF(&tmp, top, powerbuf, wvalue,
+                                            window))
             goto err;
 
         /*
@@ -833,8 +872,8 @@ int BN_mod_exp_mont_consttime(BIGNUM *rr, const BIGNUM *a, const BIGNUM *p,
             /*
              * Fetch the appropriate pre-computed value from the pre-buf
              */
-            if (!MOD_EXP_CTIME_COPY_FROM_PREBUF
-                (&am, top, powerbuf, wvalue, numPowers))
+            if (!MOD_EXP_CTIME_COPY_FROM_PREBUF(&am, top, powerbuf, wvalue,
+                                                window))
                 goto err;
 
             /* Multiply the result into the intermediate result */
diff --git a/crypto/openssl/crypto/bn/bn_print.c b/crypto/openssl/crypto/bn/bn_print.c
index ab10b957ba27..bfa31efc5621 100644
--- a/crypto/openssl/crypto/bn/bn_print.c
+++ b/crypto/openssl/crypto/bn/bn_print.c
@@ -58,6 +58,7 @@
 
 #include <stdio.h>
 #include <ctype.h>
+#include <limits.h>
 #include "cryptlib.h"
 #include <openssl/buffer.h>
 #include "bn_lcl.h"
@@ -189,7 +190,11 @@ int BN_hex2bn(BIGNUM **bn, const char *a)
         a++;
     }
 
-    for (i = 0; isxdigit((unsigned char)a[i]); i++) ;
+    for (i = 0; i <= (INT_MAX/4) && isxdigit((unsigned char)a[i]); i++)
+        continue;
+
+    if (i > INT_MAX/4)
+        goto err;
 
     num = i + neg;
     if (bn == NULL)
@@ -204,7 +209,7 @@ int BN_hex2bn(BIGNUM **bn, const char *a)
         BN_zero(ret);
     }
 
-    /* i is the number of hex digests; */
+    /* i is the number of hex digits */
     if (bn_expand(ret, i * 4) == NULL)
         goto err;
 
@@ -260,7 +265,11 @@ int BN_dec2bn(BIGNUM **bn, const char *a)
         a++;
     }
 
-    for (i = 0; isdigit((unsigned char)a[i]); i++) ;
+    for (i = 0; i <= (INT_MAX/4) && isdigit((unsigned char)a[i]); i++)
+        continue;
+
+    if (i > INT_MAX/4)
+        goto err;
 
     num = i + neg;
     if (bn == NULL)
@@ -278,7 +287,7 @@ int BN_dec2bn(BIGNUM **bn, const char *a)
         BN_zero(ret);
     }
 
-    /* i is the number of digests, a bit of an over expand; */
+    /* i is the number of digits, a bit of an over expand */
     if (bn_expand(ret, i * 4) == NULL)
         goto err;
 
diff --git a/crypto/openssl/crypto/dsa/dsa_ameth.c b/crypto/openssl/crypto/dsa/dsa_ameth.c
index a2840eaed095..d94b377b2fb3 100644
--- a/crypto/openssl/crypto/dsa/dsa_ameth.c
+++ b/crypto/openssl/crypto/dsa/dsa_ameth.c
@@ -191,6 +191,8 @@ static int dsa_priv_decode(EVP_PKEY *pkey, PKCS8_PRIV_KEY_INFO *p8)
     STACK_OF(ASN1_TYPE) *ndsa = NULL;
     DSA *dsa = NULL;
 
+    int ret = 0;
+
     if (!PKCS8_pkey_get0(NULL, &p, &pklen, &palg, p8))
         return 0;
     X509_ALGOR_get0(NULL, &ptype, &pval, palg);
@@ -262,23 +264,21 @@ static int dsa_priv_decode(EVP_PKEY *pkey, PKCS8_PRIV_KEY_INFO *p8)
     }
 
     EVP_PKEY_assign_DSA(pkey, dsa);
-    BN_CTX_free(ctx);
-    if (ndsa)
-        sk_ASN1_TYPE_pop_free(ndsa, ASN1_TYPE_free);
-    else
-        ASN1_STRING_clear_free(privkey);
 
-    return 1;
+    ret = 1;
+    goto done;
 
  decerr:
     DSAerr(DSA_F_DSA_PRIV_DECODE, EVP_R_DECODE_ERROR);
  dsaerr:
+    DSA_free(dsa);
+ done:
     BN_CTX_free(ctx);
-    if (privkey)
+    if (ndsa)
+        sk_ASN1_TYPE_pop_free(ndsa, ASN1_TYPE_free);
+    else
         ASN1_STRING_clear_free(privkey);
-    sk_ASN1_TYPE_pop_free(ndsa, ASN1_TYPE_free);
-    DSA_free(dsa);
-    return 0;
+    return ret;
 }
 
 static int dsa_priv_encode(PKCS8_PRIV_KEY_INFO *p8, const EVP_PKEY *pkey)
diff --git a/crypto/openssl/crypto/perlasm/x86_64-xlate.pl b/crypto/openssl/crypto/perlasm/x86_64-xlate.pl
index 56d9b64b6fb3..b262b8dae909 100755
--- a/crypto/openssl/crypto/perlasm/x86_64-xlate.pl
+++ b/crypto/openssl/crypto/perlasm/x86_64-xlate.pl
@@ -121,7 +121,7 @@ my %globals;
 		$self->{sz} = "";
 	    } elsif ($self->{op} =~ /^v/) { # VEX
 		$self->{sz} = "";
-	    } elsif ($self->{op} =~ /movq/ && $line =~ /%xmm/) {
+	    } elsif ($self->{op} =~ /mov[dq]/ && $line =~ /%xmm/) {
 		$self->{sz} = "";
 	    } elsif ($self->{op} =~ /([a-z]{3,})([qlwb])$/) {
 		$self->{op} = $1;
diff --git a/crypto/openssl/crypto/srp/srp.h b/crypto/openssl/crypto/srp/srp.h
index d072536fec9b..028892a1ff5e 100644
--- a/crypto/openssl/crypto/srp/srp.h
+++ b/crypto/openssl/crypto/srp/srp.h
@@ -82,16 +82,21 @@ typedef struct SRP_gN_cache_st {
 DECLARE_STACK_OF(SRP_gN_cache)
 
 typedef struct SRP_user_pwd_st {
+    /* Owned by us. */
     char *id;
     BIGNUM *s;
     BIGNUM *v;
+    /* Not owned by us. */
     const BIGNUM *g;
     const BIGNUM *N;
+    /* Owned by us. */
     char *info;
 } SRP_user_pwd;
 
 DECLARE_STACK_OF(SRP_user_pwd)
 
+void SRP_user_pwd_free(SRP_user_pwd *user_pwd);
+
 typedef struct SRP_VBASE_st {
     STACK_OF(SRP_user_pwd) *users_pwd;
     STACK_OF(SRP_gN_cache) *gN_cache;
@@ -115,7 +120,12 @@ DECLARE_STACK_OF(SRP_gN)
 SRP_VBASE *SRP_VBASE_new(char *seed_key);
 int SRP_VBASE_free(SRP_VBASE *vb);
 int SRP_VBASE_init(SRP_VBASE *vb, char *verifier_file);
+
+/* This method ignores the configured seed and fails for an unknown user. */
 SRP_user_pwd *SRP_VBASE_get_by_user(SRP_VBASE *vb, char *username);
+/* NOTE: unlike in SRP_VBASE_get_by_user, caller owns the returned pointer.*/
+SRP_user_pwd *SRP_VBASE_get1_by_user(SRP_VBASE *vb, char *username);
+
 char *SRP_create_verifier(const char *user, const char *pass, char **salt,
                           char **verifier, const char *N, const char *g);
 int SRP_create_verifier_BN(const char *user, const char *pass, BIGNUM **salt,
diff --git a/crypto/openssl/crypto/srp/srp_vfy.c b/crypto/openssl/crypto/srp/srp_vfy.c
index 50f75d7e4c9f..20989ef0fede 100644
--- a/crypto/openssl/crypto/srp/srp_vfy.c
+++ b/crypto/openssl/crypto/srp/srp_vfy.c
@@ -185,7 +185,7 @@ static char *t_tob64(char *dst, const unsigned char *src, int size)
     return olddst;
 }
 
-static void SRP_user_pwd_free(SRP_user_pwd *user_pwd)
+void SRP_user_pwd_free(SRP_user_pwd *user_pwd)
 {
     if (user_pwd == NULL)
         return;
@@ -247,6 +247,24 @@ static int SRP_user_pwd_set_sv_BN(SRP_user_pwd *vinfo, BIGNUM *s, BIGNUM *v)
     return (vinfo->s != NULL && vinfo->v != NULL);
 }
 
+static SRP_user_pwd *srp_user_pwd_dup(SRP_user_pwd *src)
+{
+    SRP_user_pwd *ret;
+
+    if (src == NULL)
+        return NULL;
+    if ((ret = SRP_user_pwd_new()) == NULL)
+        return NULL;
+
+    SRP_user_pwd_set_gN(ret, src->g, src->N);
+    if (!SRP_user_pwd_set_ids(ret, src->id, src->info)
+        || !SRP_user_pwd_set_sv_BN(ret, BN_dup(src->s), BN_dup(src->v))) {
+            SRP_user_pwd_free(ret);
+            return NULL;
+    }
+    return ret;
+}
+
 SRP_VBASE *SRP_VBASE_new(char *seed_key)
 {
     SRP_VBASE *vb = (SRP_VBASE *)OPENSSL_malloc(sizeof(SRP_VBASE));
@@ -468,21 +486,50 @@ int SRP_VBASE_init(SRP_VBASE *vb, char *verifier_file)
 
 }
 
-SRP_user_pwd *SRP_VBASE_get_by_user(SRP_VBASE *vb, char *username)
+static SRP_user_pwd *find_user(SRP_VBASE *vb, char *username)
 {
     int i;
     SRP_user_pwd *user;
-    unsigned char digv[SHA_DIGEST_LENGTH];
-    unsigned char digs[SHA_DIGEST_LENGTH];
-    EVP_MD_CTX ctxt;
 
     if (vb == NULL)
         return NULL;
+
     for (i = 0; i < sk_SRP_user_pwd_num(vb->users_pwd); i++) {
         user = sk_SRP_user_pwd_value(vb->users_pwd, i);
         if (strcmp(user->id, username) == 0)
             return user;
     }
+
+    return NULL;
+}
+
+/*
+ * This method ignores the configured seed and fails for an unknown user.
+ * Ownership of the returned pointer is not released to the caller.
+ * In other words, caller must not free the result.
+ */
+SRP_user_pwd *SRP_VBASE_get_by_user(SRP_VBASE *vb, char *username)
+{
+    return find_user(vb, username);
+}
+
+/*
+ * Ownership of the returned pointer is released to the caller.
+ * In other words, caller must free the result once done.
+ */
+SRP_user_pwd *SRP_VBASE_get1_by_user(SRP_VBASE *vb, char *username)
+{
+    SRP_user_pwd *user;
+    unsigned char digv[SHA_DIGEST_LENGTH];
+    unsigned char digs[SHA_DIGEST_LENGTH];
+    EVP_MD_CTX ctxt;
+
+    if (vb == NULL)
+        return NULL;
+
+    if ((user = find_user(vb, username)) != NULL)
+        return srp_user_pwd_dup(user);
+
     if ((vb->seed_key == NULL) ||
         (vb->default_g == NULL) || (vb->default_N == NULL))
         return NULL;
diff --git a/crypto/openssl/ssl/s2_lib.c b/crypto/openssl/ssl/s2_lib.c
index 7e3674a685fb..82c173154980 100644
--- a/crypto/openssl/ssl/s2_lib.c
+++ b/crypto/openssl/ssl/s2_lib.c
@@ -156,6 +156,7 @@ OPENSSL_GLOBAL const SSL_CIPHER ssl2_ciphers[] = {
      128,
      },
 
+# if 0
 /* RC4_128_EXPORT40_WITH_MD5 */
     {
      1,
@@ -171,6 +172,7 @@ OPENSSL_GLOBAL const SSL_CIPHER ssl2_ciphers[] = {
      40,
      128,
      },
+# endif
 
 /* RC2_128_CBC_WITH_MD5 */
     {
@@ -188,6 +190,7 @@ OPENSSL_GLOBAL const SSL_CIPHER ssl2_ciphers[] = {
      128,
      },
 
+# if 0
 /* RC2_128_CBC_EXPORT40_WITH_MD5 */
     {
      1,
@@ -203,6 +206,7 @@ OPENSSL_GLOBAL const SSL_CIPHER ssl2_ciphers[] = {
      40,
      128,
      },
+# endif
 
 # ifndef OPENSSL_NO_IDEA
 /* IDEA_128_CBC_WITH_MD5 */
@@ -222,6 +226,7 @@ OPENSSL_GLOBAL const SSL_CIPHER ssl2_ciphers[] = {
      },
 # endif
 
+# if 0
 /* DES_64_CBC_WITH_MD5 */
     {
      1,
@@ -237,6 +242,7 @@ OPENSSL_GLOBAL const SSL_CIPHER ssl2_ciphers[] = {
      56,
      56,
      },
+# endif
 
 /* DES_192_EDE3_CBC_WITH_MD5 */
     {
diff --git a/crypto/openssl/ssl/s3_lib.c b/crypto/openssl/ssl/s3_lib.c
index de917d3f836f..f0682ad965d4 100644
--- a/crypto/openssl/ssl/s3_lib.c
+++ b/crypto/openssl/ssl/s3_lib.c
@@ -203,6 +203,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      },
 
 /* Cipher 03 */
+#ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_RSA_RC4_40_MD5,
@@ -217,6 +218,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      40,
      128,
      },
+#endif
 
 /* Cipher 04 */
     {
@@ -251,6 +253,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      },
 
 /* Cipher 06 */
+#ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_RSA_RC2_40_MD5,
@@ -265,6 +268,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      40,
      128,
      },
+#endif
 
 /* Cipher 07 */
 #ifndef OPENSSL_NO_IDEA
@@ -285,6 +289,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
 #endif
 
 /* Cipher 08 */
+#ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_RSA_DES_40_CBC_SHA,
@@ -299,8 +304,10 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      40,
      56,
      },
+#endif
 
 /* Cipher 09 */
+#ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_RSA_DES_64_CBC_SHA,
@@ -315,6 +322,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      56,
      56,
      },
+#endif
 
 /* Cipher 0A */
     {
@@ -334,6 +342,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
 
 /* The DH ciphers */
 /* Cipher 0B */
+#ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      0,
      SSL3_TXT_DH_DSS_DES_40_CBC_SHA,
@@ -348,8 +357,10 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      40,
      56,
      },
+#endif
 
 /* Cipher 0C */
+#ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      0,                         /* not implemented (non-ephemeral DH) */
      SSL3_TXT_DH_DSS_DES_64_CBC_SHA,
@@ -364,6 +375,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      56,
      56,
      },
+#endif
 
 /* Cipher 0D */
     {
@@ -382,6 +394,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      },
 
 /* Cipher 0E */
+#ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      0,                         /* not implemented (non-ephemeral DH) */
      SSL3_TXT_DH_RSA_DES_40_CBC_SHA,
@@ -396,8 +409,10 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      40,
      56,
      },
+#endif
 
 /* Cipher 0F */
+#ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      0,                         /* not implemented (non-ephemeral DH) */
      SSL3_TXT_DH_RSA_DES_64_CBC_SHA,
@@ -412,6 +427,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      56,
      56,
      },
+#endif
 
 /* Cipher 10 */
     {
@@ -431,6 +447,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
 
 /* The Ephemeral DH ciphers */
 /* Cipher 11 */
+#ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_EDH_DSS_DES_40_CBC_SHA,
@@ -445,8 +462,10 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      40,
      56,
      },
+#endif
 
 /* Cipher 12 */
+#ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_EDH_DSS_DES_64_CBC_SHA,
@@ -461,6 +480,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      56,
      56,
      },
+#endif
 
 /* Cipher 13 */
     {
@@ -479,6 +499,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      },
 
 /* Cipher 14 */
+#ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_EDH_RSA_DES_40_CBC_SHA,
@@ -493,8 +514,10 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      40,
      56,
      },
+#endif
 
 /* Cipher 15 */
+#ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_EDH_RSA_DES_64_CBC_SHA,
@@ -509,6 +532,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      56,
      56,
      },
+#endif
 
 /* Cipher 16 */
     {
@@ -527,6 +551,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      },
 
 /* Cipher 17 */
+#ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_ADH_RC4_40_MD5,
@@ -541,6 +566,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      40,
      128,
      },
+#endif
 
 /* Cipher 18 */
     {
@@ -559,6 +585,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      },
 
 /* Cipher 19 */
+#ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_ADH_DES_40_CBC_SHA,
@@ -573,8 +600,10 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      40,
      128,
      },
+#endif
 
 /* Cipher 1A */
+#ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_ADH_DES_64_CBC_SHA,
@@ -589,6 +618,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      56,
      56,
      },
+#endif
 
 /* Cipher 1B */
     {
@@ -660,6 +690,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
 #ifndef OPENSSL_NO_KRB5
 /* The Kerberos ciphers*/
 /* Cipher 1E */
+# ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_KRB5_DES_64_CBC_SHA,
@@ -674,6 +705,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      56,
      56,
      },
+# endif
 
 /* Cipher 1F */
     {
@@ -724,6 +756,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      },
 
 /* Cipher 22 */
+# ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_KRB5_DES_64_CBC_MD5,
@@ -738,6 +771,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      56,
      56,
      },
+# endif
 
 /* Cipher 23 */
     {
@@ -788,6 +822,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      },
 
 /* Cipher 26 */
+# ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_KRB5_DES_40_CBC_SHA,
@@ -802,8 +837,10 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      40,
      56,
      },
+# endif
 
 /* Cipher 27 */
+# ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_KRB5_RC2_40_CBC_SHA,
@@ -818,8 +855,10 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      40,
      128,
      },
+# endif
 
 /* Cipher 28 */
+# ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_KRB5_RC4_40_SHA,
@@ -834,8 +873,10 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      40,
      128,
      },
+# endif
 
 /* Cipher 29 */
+# ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_KRB5_DES_40_CBC_MD5,
@@ -850,8 +891,10 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      40,
      56,
      },
+# endif
 
 /* Cipher 2A */
+# ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_KRB5_RC2_40_CBC_MD5,
@@ -866,8 +909,10 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      40,
      128,
      },
+# endif
 
 /* Cipher 2B */
+# ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      SSL3_TXT_KRB5_RC4_40_MD5,
@@ -882,6 +927,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      40,
      128,
      },
+# endif
 #endif                          /* OPENSSL_NO_KRB5 */
 
 /* New AES ciphersuites */
@@ -1305,6 +1351,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
 # endif
 
     /* Cipher 62 */
+# ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      TLS1_TXT_RSA_EXPORT1024_WITH_DES_CBC_SHA,
@@ -1319,8 +1366,10 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      56,
      56,
      },
+# endif
 
     /* Cipher 63 */
+# ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      TLS1_TXT_DHE_DSS_EXPORT1024_WITH_DES_CBC_SHA,
@@ -1335,8 +1384,10 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      56,
      56,
      },
+# endif
 
     /* Cipher 64 */
+# ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      TLS1_TXT_RSA_EXPORT1024_WITH_RC4_56_SHA,
@@ -1351,8 +1402,10 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      56,
      128,
      },
+# endif
 
     /* Cipher 65 */
+# ifndef OPENSSL_NO_WEAK_SSL_CIPHERS
     {
      1,
      TLS1_TXT_DHE_DSS_EXPORT1024_WITH_RC4_56_SHA,
@@ -1367,6 +1420,7 @@ OPENSSL_GLOBAL SSL_CIPHER ssl3_ciphers[] = {
      56,
      128,
      },
+# endif
 
     /* Cipher 66 */
     {
diff --git a/crypto/openssl/ssl/ssl_lib.c b/crypto/openssl/ssl/ssl_lib.c
index e11746a69508..e8558d59bcce 100644
--- a/crypto/openssl/ssl/ssl_lib.c
+++ b/crypto/openssl/ssl/ssl_lib.c
@@ -1896,6 +1896,13 @@ SSL_CTX *SSL_CTX_new(const SSL_METHOD *meth)
      */
     ret->options |= SSL_OP_LEGACY_SERVER_CONNECT;
 
+    /*
+     * Disable SSLv2 by default, callers that want to enable SSLv2 will have to
+     * explicitly clear this option via either of SSL_CTX_clear_options() or
+     * SSL_clear_options().
+     */
+    ret->options |= SSL_OP_NO_SSLv2;
+
     return (ret);
  err:
     SSLerr(SSL_F_SSL_CTX_NEW, ERR_R_MALLOC_FAILURE);
diff --git a/crypto/openssl/util/libeay.num b/crypto/openssl/util/libeay.num
index b594caf2cf79..a83c3bedac3c 100644
--- a/crypto/openssl/util/libeay.num
+++ b/crypto/openssl/util/libeay.num
@@ -1807,6 +1807,8 @@ ASN1_UTCTIME_get                        2350	NOEXIST::FUNCTION:
 X509_REQ_digest                         2362	EXIST::FUNCTION:EVP
 X509_CRL_digest                         2391	EXIST::FUNCTION:EVP
 ASN1_STRING_clear_free                  2392	EXIST::FUNCTION:
+SRP_VBASE_get1_by_user                  2393	EXIST::FUNCTION:SRP
+SRP_user_pwd_free                       2394	EXIST::FUNCTION:SRP
 d2i_ASN1_SET_OF_PKCS7                   2397	NOEXIST::FUNCTION:
 X509_ALGOR_cmp                          2398	EXIST::FUNCTION:
 EVP_CIPHER_CTX_set_key_length           2399	EXIST::FUNCTION:
diff --git a/secure/lib/libcrypto/amd64/x86_64-mont5.S b/secure/lib/libcrypto/amd64/x86_64-mont5.S
index b0b34423b0b1..5820e12ec3a0 100644
--- a/secure/lib/libcrypto/amd64/x86_64-mont5.S
+++ b/secure/lib/libcrypto/amd64/x86_64-mont5.S
@@ -14,47 +14,153 @@ bn_mul_mont_gather5:
 .align	16
 .Lmul_enter:
 	movl	%r9d,%r9d
-	movl	8(%rsp),%r10d
+	movd	8(%rsp),%xmm5
+	leaq	.Linc(%rip),%r10
 	pushq	%rbx
 	pushq	%rbp
 	pushq	%r12
 	pushq	%r13
 	pushq	%r14
 	pushq	%r15
+
+.Lmul_alloca:
 	movq	%rsp,%rax
 	leaq	2(%r9),%r11
 	negq	%r11
-	leaq	(%rsp,%r11,8),%rsp
+	leaq	-264(%rsp,%r11,8),%rsp
 	andq	$-1024,%rsp
 
 	movq	%rax,8(%rsp,%r9,8)
 .Lmul_body:
-	movq	%rdx,%r12
-	movq	%r10,%r11
-	shrq	$3,%r10
-	andq	$7,%r11
-	notq	%r10
-	leaq	.Lmagic_masks(%rip),%rax
-	andq	$3,%r10
-	leaq	96(%r12,%r11,8),%r12
-	movq	0(%rax,%r10,8),%xmm4
-	movq	8(%rax,%r10,8),%xmm5
-	movq	16(%rax,%r10,8),%xmm6
-	movq	24(%rax,%r10,8),%xmm7
-
-	movq	-96(%r12),%xmm0
-	movq	-32(%r12),%xmm1
-	pand	%xmm4,%xmm0
-	movq	32(%r12),%xmm2
-	pand	%xmm5,%xmm1
-	movq	96(%r12),%xmm3
-	pand	%xmm6,%xmm2
-	por	%xmm1,%xmm0
-	pand	%xmm7,%xmm3
+	leaq	128(%rdx),%r12
+	movdqa	0(%r10),%xmm0
+	movdqa	16(%r10),%xmm1
+	leaq	24-112(%rsp,%r9,8),%r10
+	andq	$-16,%r10
+
+	pshufd	$0,%xmm5,%xmm5
+	movdqa	%xmm1,%xmm4
+	movdqa	%xmm1,%xmm2
+	paddd	%xmm0,%xmm1
+	pcmpeqd	%xmm5,%xmm0
+.byte	0x67
+	movdqa	%xmm4,%xmm3
+	paddd	%xmm1,%xmm2
+	pcmpeqd	%xmm5,%xmm1
+	movdqa	%xmm0,112(%r10)
+	movdqa	%xmm4,%xmm0
+
+	paddd	%xmm2,%xmm3
+	pcmpeqd	%xmm5,%xmm2
+	movdqa	%xmm1,128(%r10)
+	movdqa	%xmm4,%xmm1
+
+	paddd	%xmm3,%xmm0
+	pcmpeqd	%xmm5,%xmm3
+	movdqa	%xmm2,144(%r10)
+	movdqa	%xmm4,%xmm2
+
+	paddd	%xmm0,%xmm1
+	pcmpeqd	%xmm5,%xmm0
+	movdqa	%xmm3,160(%r10)
+	movdqa	%xmm4,%xmm3
+	paddd	%xmm1,%xmm2
+	pcmpeqd	%xmm5,%xmm1
+	movdqa	%xmm0,176(%r10)
+	movdqa	%xmm4,%xmm0
+
+	paddd	%xmm2,%xmm3
+	pcmpeqd	%xmm5,%xmm2
+	movdqa	%xmm1,192(%r10)
+	movdqa	%xmm4,%xmm1
+
+	paddd	%xmm3,%xmm0
+	pcmpeqd	%xmm5,%xmm3
+	movdqa	%xmm2,208(%r10)
+	movdqa	%xmm4,%xmm2
+
+	paddd	%xmm0,%xmm1
+	pcmpeqd	%xmm5,%xmm0
+	movdqa	%xmm3,224(%r10)
+	movdqa	%xmm4,%xmm3
+	paddd	%xmm1,%xmm2
+	pcmpeqd	%xmm5,%xmm1
+	movdqa	%xmm0,240(%r10)
+	movdqa	%xmm4,%xmm0
+
+	paddd	%xmm2,%xmm3
+	pcmpeqd	%xmm5,%xmm2
+	movdqa	%xmm1,256(%r10)
+	movdqa	%xmm4,%xmm1
+
+	paddd	%xmm3,%xmm0
+	pcmpeqd	%xmm5,%xmm3
+	movdqa	%xmm2,272(%r10)
+	movdqa	%xmm4,%xmm2
+
+	paddd	%xmm0,%xmm1
+	pcmpeqd	%xmm5,%xmm0
+	movdqa	%xmm3,288(%r10)
+	movdqa	%xmm4,%xmm3
+	paddd	%xmm1,%xmm2
+	pcmpeqd	%xmm5,%xmm1
+	movdqa	%xmm0,304(%r10)
+
+	paddd	%xmm2,%xmm3
+.byte	0x67
+	pcmpeqd	%xmm5,%xmm2
+	movdqa	%xmm1,320(%r10)
+
+	pcmpeqd	%xmm5,%xmm3
+	movdqa	%xmm2,336(%r10)
+	pand	64(%r12),%xmm0
+
+	pand	80(%r12),%xmm1
+	pand	96(%r12),%xmm2
+	movdqa	%xmm3,352(%r10)
+	pand	112(%r12),%xmm3
+	por	%xmm2,%xmm0
+	por	%xmm3,%xmm1
+	movdqa	-128(%r12),%xmm4
+	movdqa	-112(%r12),%xmm5
+	movdqa	-96(%r12),%xmm2
+	pand	112(%r10),%xmm4
+	movdqa	-80(%r12),%xmm3
+	pand	128(%r10),%xmm5
+	por	%xmm4,%xmm0
+	pand	144(%r10),%xmm2
+	por	%xmm5,%xmm1
+	pand	160(%r10),%xmm3
 	por	%xmm2,%xmm0
+	por	%xmm3,%xmm1
+	movdqa	-64(%r12),%xmm4
+	movdqa	-48(%r12),%xmm5
+	movdqa	-32(%r12),%xmm2
+	pand	176(%r10),%xmm4
+	movdqa	-16(%r12),%xmm3
+	pand	192(%r10),%xmm5
+	por	%xmm4,%xmm0
+	pand	208(%r10),%xmm2
+	por	%xmm5,%xmm1
+	pand	224(%r10),%xmm3
+	por	%xmm2,%xmm0
+	por	%xmm3,%xmm1
+	movdqa	0(%r12),%xmm4
+	movdqa	16(%r12),%xmm5
+	movdqa	32(%r12),%xmm2
+	pand	240(%r10),%xmm4
+	movdqa	48(%r12),%xmm3
+	pand	256(%r10),%xmm5
+	por	%xmm4,%xmm0
+	pand	272(%r10),%xmm2
+	por	%xmm5,%xmm1
+	pand	288(%r10),%xmm3
+	por	%xmm2,%xmm0
+	por	%xmm3,%xmm1
+	por	%xmm1,%xmm0
+	pshufd	$78,%xmm0,%xmm1
+	por	%xmm1,%xmm0
 	leaq	256(%r12),%r12
-	por	%xmm3,%xmm0
-
 .byte	102,72,15,126,195
 
 	movq	(%r8),%r8
@@ -63,29 +169,14 @@ bn_mul_mont_gather5:
 	xorq	%r14,%r14
 	xorq	%r15,%r15
 
-	movq	-96(%r12),%xmm0
-	movq	-32(%r12),%xmm1
-	pand	%xmm4,%xmm0
-	movq	32(%r12),%xmm2
-	pand	%xmm5,%xmm1
-
 	movq	%r8,%rbp
 	mulq	%rbx
 	movq	%rax,%r10
 	movq	(%rcx),%rax
 
-	movq	96(%r12),%xmm3
-	pand	%xmm6,%xmm2
-	por	%xmm1,%xmm0
-	pand	%xmm7,%xmm3
-
 	imulq	%r10,%rbp
 	movq	%rdx,%r11
 
-	por	%xmm2,%xmm0
-	leaq	256(%r12),%r12
-	por	%xmm3,%xmm0
-
 	mulq	%rbp
 	addq	%rax,%r10
 	movq	8(%rsi),%rax
@@ -118,8 +209,6 @@ bn_mul_mont_gather5:
 	cmpq	%r9,%r15
 	jne	.L1st
 
-.byte	102,72,15,126,195
-
 	addq	%rax,%r13
 	movq	(%rsi),%rax
 	adcq	$0,%rdx
@@ -139,33 +228,76 @@ bn_mul_mont_gather5:
 	jmp	.Louter
 .align	16
 .Louter:
+	leaq	24+128(%rsp,%r9,8),%rdx
+	andq	$-16,%rdx
+	pxor	%xmm4,%xmm4
+	pxor	%xmm5,%xmm5
+	movdqa	-128(%r12),%xmm0
+	movdqa	-112(%r12),%xmm1
+	movdqa	-96(%r12),%xmm2
+	movdqa	-80(%r12),%xmm3
+	pand	-128(%rdx),%xmm0
+	pand	-112(%rdx),%xmm1
+	por	%xmm0,%xmm4
+	pand	-96(%rdx),%xmm2
+	por	%xmm1,%xmm5
+	pand	-80(%rdx),%xmm3
+	por	%xmm2,%xmm4
+	por	%xmm3,%xmm5
+	movdqa	-64(%r12),%xmm0
+	movdqa	-48(%r12),%xmm1
+	movdqa	-32(%r12),%xmm2
+	movdqa	-16(%r12),%xmm3
+	pand	-64(%rdx),%xmm0
+	pand	-48(%rdx),%xmm1
+	por	%xmm0,%xmm4
+	pand	-32(%rdx),%xmm2
+	por	%xmm1,%xmm5
+	pand	-16(%rdx),%xmm3
+	por	%xmm2,%xmm4
+	por	%xmm3,%xmm5
+	movdqa	0(%r12),%xmm0
+	movdqa	16(%r12),%xmm1
+	movdqa	32(%r12),%xmm2
+	movdqa	48(%r12),%xmm3
+	pand	0(%rdx),%xmm0
+	pand	16(%rdx),%xmm1
+	por	%xmm0,%xmm4
+	pand	32(%rdx),%xmm2
+	por	%xmm1,%xmm5
+	pand	48(%rdx),%xmm3
+	por	%xmm2,%xmm4
+	por	%xmm3,%xmm5
+	movdqa	64(%r12),%xmm0
+	movdqa	80(%r12),%xmm1
+	movdqa	96(%r12),%xmm2
+	movdqa	112(%r12),%xmm3
+	pand	64(%rdx),%xmm0
+	pand	80(%rdx),%xmm1
+	por	%xmm0,%xmm4
+	pand	96(%rdx),%xmm2
+	por	%xmm1,%xmm5
+	pand	112(%rdx),%xmm3
+	por	%xmm2,%xmm4
+	por	%xmm3,%xmm5
+	por	%xmm5,%xmm4
+	pshufd	$78,%xmm4,%xmm0
+	por	%xmm4,%xmm0
+	leaq	256(%r12),%r12
+.byte	102,72,15,126,195
+
 	xorq	%r15,%r15
 	movq	%r8,%rbp
 	movq	(%rsp),%r10
 
-	movq	-96(%r12),%xmm0
-	movq	-32(%r12),%xmm1
-	pand	%xmm4,%xmm0
-	movq	32(%r12),%xmm2
-	pand	%xmm5,%xmm1
-
 	mulq	%rbx
 	addq	%rax,%r10
 	movq	(%rcx),%rax
 	adcq	$0,%rdx
 
-	movq	96(%r12),%xmm3
-	pand	%xmm6,%xmm2
-	por	%xmm1,%xmm0
-	pand	%xmm7,%xmm3
-
 	imulq	%r10,%rbp
 	movq	%rdx,%r11
 
-	por	%xmm2,%xmm0
-	leaq	256(%r12),%r12
-	por	%xmm3,%xmm0
-
 	mulq	%rbp
 	addq	%rax,%r10
 	movq	8(%rsi),%rax
@@ -201,8 +333,6 @@ bn_mul_mont_gather5:
 	cmpq	%r9,%r15
 	jne	.Linner
 
-.byte	102,72,15,126,195
-
 	addq	%rax,%r13
 	movq	(%rsi),%rax
 	adcq	$0,%rdx
@@ -256,6 +386,7 @@ bn_mul_mont_gather5:
 
 	movq	8(%rsp,%r9,8),%rsi
 	movq	$1,%rax
+
 	movq	(%rsi),%r15
 	movq	8(%rsi),%r14
 	movq	16(%rsi),%r13
@@ -271,78 +402,170 @@ bn_mul_mont_gather5:
 bn_mul4x_mont_gather5:
 .Lmul4x_enter:
 	movl	%r9d,%r9d
-	movl	8(%rsp),%r10d
+	movd	8(%rsp),%xmm5
+	leaq	.Linc(%rip),%r10
 	pushq	%rbx
 	pushq	%rbp
 	pushq	%r12
 	pushq	%r13
 	pushq	%r14
 	pushq	%r15
+
+.Lmul4x_alloca:
 	movq	%rsp,%rax
 	leaq	4(%r9),%r11
 	negq	%r11
-	leaq	(%rsp,%r11,8),%rsp
+	leaq	-256(%rsp,%r11,8),%rsp
 	andq	$-1024,%rsp
 
 	movq	%rax,8(%rsp,%r9,8)
 .Lmul4x_body:
 	movq	%rdi,16(%rsp,%r9,8)
-	movq	%rdx,%r12
-	movq	%r10,%r11
-	shrq	$3,%r10
-	andq	$7,%r11
-	notq	%r10
-	leaq	.Lmagic_masks(%rip),%rax
-	andq	$3,%r10
-	leaq	96(%r12,%r11,8),%r12
-	movq	0(%rax,%r10,8),%xmm4
-	movq	8(%rax,%r10,8),%xmm5
-	movq	16(%rax,%r10,8),%xmm6
-	movq	24(%rax,%r10,8),%xmm7
-
-	movq	-96(%r12),%xmm0
-	movq	-32(%r12),%xmm1
-	pand	%xmm4,%xmm0
-	movq	32(%r12),%xmm2
-	pand	%xmm5,%xmm1
-	movq	96(%r12),%xmm3
-	pand	%xmm6,%xmm2
-	por	%xmm1,%xmm0
-	pand	%xmm7,%xmm3
+	leaq	128(%rdx),%r12
+	movdqa	0(%r10),%xmm0
+	movdqa	16(%r10),%xmm1
+	leaq	32-112(%rsp,%r9,8),%r10
+
+	pshufd	$0,%xmm5,%xmm5
+	movdqa	%xmm1,%xmm4
+.byte	0x67,0x67
+	movdqa	%xmm1,%xmm2
+	paddd	%xmm0,%xmm1
+	pcmpeqd	%xmm5,%xmm0
+.byte	0x67
+	movdqa	%xmm4,%xmm3
+	paddd	%xmm1,%xmm2
+	pcmpeqd	%xmm5,%xmm1
+	movdqa	%xmm0,112(%r10)
+	movdqa	%xmm4,%xmm0
+
+	paddd	%xmm2,%xmm3
+	pcmpeqd	%xmm5,%xmm2
+	movdqa	%xmm1,128(%r10)
+	movdqa	%xmm4,%xmm1
+
+	paddd	%xmm3,%xmm0
+	pcmpeqd	%xmm5,%xmm3
+	movdqa	%xmm2,144(%r10)
+	movdqa	%xmm4,%xmm2
+
+	paddd	%xmm0,%xmm1
+	pcmpeqd	%xmm5,%xmm0
+	movdqa	%xmm3,160(%r10)
+	movdqa	%xmm4,%xmm3
+	paddd	%xmm1,%xmm2
+	pcmpeqd	%xmm5,%xmm1
+	movdqa	%xmm0,176(%r10)
+	movdqa	%xmm4,%xmm0
+
+	paddd	%xmm2,%xmm3
+	pcmpeqd	%xmm5,%xmm2
+	movdqa	%xmm1,192(%r10)
+	movdqa	%xmm4,%xmm1
+
+	paddd	%xmm3,%xmm0
+	pcmpeqd	%xmm5,%xmm3
+	movdqa	%xmm2,208(%r10)
+	movdqa	%xmm4,%xmm2
+
+	paddd	%xmm0,%xmm1
+	pcmpeqd	%xmm5,%xmm0
+	movdqa	%xmm3,224(%r10)
+	movdqa	%xmm4,%xmm3
+	paddd	%xmm1,%xmm2
+	pcmpeqd	%xmm5,%xmm1
+	movdqa	%xmm0,240(%r10)
+	movdqa	%xmm4,%xmm0
+
+	paddd	%xmm2,%xmm3
+	pcmpeqd	%xmm5,%xmm2
+	movdqa	%xmm1,256(%r10)
+	movdqa	%xmm4,%xmm1
+
+	paddd	%xmm3,%xmm0
+	pcmpeqd	%xmm5,%xmm3
+	movdqa	%xmm2,272(%r10)
+	movdqa	%xmm4,%xmm2
+
+	paddd	%xmm0,%xmm1
+	pcmpeqd	%xmm5,%xmm0
+	movdqa	%xmm3,288(%r10)
+	movdqa	%xmm4,%xmm3
+	paddd	%xmm1,%xmm2
+	pcmpeqd	%xmm5,%xmm1
+	movdqa	%xmm0,304(%r10)
+
+	paddd	%xmm2,%xmm3
+.byte	0x67
+	pcmpeqd	%xmm5,%xmm2
+	movdqa	%xmm1,320(%r10)
+
+	pcmpeqd	%xmm5,%xmm3
+	movdqa	%xmm2,336(%r10)
+	pand	64(%r12),%xmm0
+
+	pand	80(%r12),%xmm1
+	pand	96(%r12),%xmm2
+	movdqa	%xmm3,352(%r10)
+	pand	112(%r12),%xmm3
 	por	%xmm2,%xmm0
+	por	%xmm3,%xmm1
+	movdqa	-128(%r12),%xmm4
+	movdqa	-112(%r12),%xmm5
+	movdqa	-96(%r12),%xmm2
+	pand	112(%r10),%xmm4
+	movdqa	-80(%r12),%xmm3
+	pand	128(%r10),%xmm5
+	por	%xmm4,%xmm0
+	pand	144(%r10),%xmm2
+	por	%xmm5,%xmm1
+	pand	160(%r10),%xmm3
+	por	%xmm2,%xmm0
+	por	%xmm3,%xmm1
+	movdqa	-64(%r12),%xmm4
+	movdqa	-48(%r12),%xmm5
+	movdqa	-32(%r12),%xmm2
+	pand	176(%r10),%xmm4
+	movdqa	-16(%r12),%xmm3
+	pand	192(%r10),%xmm5
+	por	%xmm4,%xmm0
+	pand	208(%r10),%xmm2
+	por	%xmm5,%xmm1
+	pand	224(%r10),%xmm3
+	por	%xmm2,%xmm0
+	por	%xmm3,%xmm1
+	movdqa	0(%r12),%xmm4
+	movdqa	16(%r12),%xmm5
+	movdqa	32(%r12),%xmm2
+	pand	240(%r10),%xmm4
+	movdqa	48(%r12),%xmm3
+	pand	256(%r10),%xmm5
+	por	%xmm4,%xmm0
+	pand	272(%r10),%xmm2
+	por	%xmm5,%xmm1
+	pand	288(%r10),%xmm3
+	por	%xmm2,%xmm0
+	por	%xmm3,%xmm1
+	por	%xmm1,%xmm0
+	pshufd	$78,%xmm0,%xmm1
+	por	%xmm1,%xmm0
 	leaq	256(%r12),%r12
-	por	%xmm3,%xmm0
-
 .byte	102,72,15,126,195
+
 	movq	(%r8),%r8
 	movq	(%rsi),%rax
 
 	xorq	%r14,%r14
 	xorq	%r15,%r15
 
-	movq	-96(%r12),%xmm0
-	movq	-32(%r12),%xmm1
-	pand	%xmm4,%xmm0
-	movq	32(%r12),%xmm2
-	pand	%xmm5,%xmm1
-
 	movq	%r8,%rbp
 	mulq	%rbx
 	movq	%rax,%r10
 	movq	(%rcx),%rax
 
-	movq	96(%r12),%xmm3
-	pand	%xmm6,%xmm2
-	por	%xmm1,%xmm0
-	pand	%xmm7,%xmm3
-
 	imulq	%r10,%rbp
 	movq	%rdx,%r11
 
-	por	%xmm2,%xmm0
-	leaq	256(%r12),%r12
-	por	%xmm3,%xmm0
-
 	mulq	%rbp
 	addq	%rax,%r10
 	movq	8(%rsi),%rax
@@ -460,8 +683,6 @@ bn_mul4x_mont_gather5:
 	movq	%rdi,-16(%rsp,%r15,8)
 	movq	%rdx,%r13
 
-.byte	102,72,15,126,195
-
 	xorq	%rdi,%rdi
 	addq	%r10,%r13
 	adcq	$0,%rdi
@@ -471,12 +692,64 @@ bn_mul4x_mont_gather5:
 	leaq	1(%r14),%r14
 .align	4
 .Louter4x:
+	leaq	32+128(%rsp,%r9,8),%rdx
+	pxor	%xmm4,%xmm4
+	pxor	%xmm5,%xmm5
+	movdqa	-128(%r12),%xmm0
+	movdqa	-112(%r12),%xmm1
+	movdqa	-96(%r12),%xmm2
+	movdqa	-80(%r12),%xmm3
+	pand	-128(%rdx),%xmm0
+	pand	-112(%rdx),%xmm1
+	por	%xmm0,%xmm4
+	pand	-96(%rdx),%xmm2
+	por	%xmm1,%xmm5
+	pand	-80(%rdx),%xmm3
+	por	%xmm2,%xmm4
+	por	%xmm3,%xmm5
+	movdqa	-64(%r12),%xmm0
+	movdqa	-48(%r12),%xmm1
+	movdqa	-32(%r12),%xmm2
+	movdqa	-16(%r12),%xmm3
+	pand	-64(%rdx),%xmm0
+	pand	-48(%rdx),%xmm1
+	por	%xmm0,%xmm4
+	pand	-32(%rdx),%xmm2
+	por	%xmm1,%xmm5
+	pand	-16(%rdx),%xmm3
+	por	%xmm2,%xmm4
+	por	%xmm3,%xmm5
+	movdqa	0(%r12),%xmm0
+	movdqa	16(%r12),%xmm1
+	movdqa	32(%r12),%xmm2
+	movdqa	48(%r12),%xmm3
+	pand	0(%rdx),%xmm0
+	pand	16(%rdx),%xmm1
+	por	%xmm0,%xmm4
+	pand	32(%rdx),%xmm2
+	por	%xmm1,%xmm5
+	pand	48(%rdx),%xmm3
+	por	%xmm2,%xmm4
+	por	%xmm3,%xmm5
+	movdqa	64(%r12),%xmm0
+	movdqa	80(%r12),%xmm1
+	movdqa	96(%r12),%xmm2
+	movdqa	112(%r12),%xmm3
+	pand	64(%rdx),%xmm0
+	pand	80(%rdx),%xmm1
+	por	%xmm0,%xmm4
+	pand	96(%rdx),%xmm2
+	por	%xmm1,%xmm5
+	pand	112(%rdx),%xmm3
+	por	%xmm2,%xmm4
+	por	%xmm3,%xmm5
+	por	%xmm5,%xmm4
+	pshufd	$78,%xmm4,%xmm0
+	por	%xmm4,%xmm0
+	leaq	256(%r12),%r12
+.byte	102,72,15,126,195
+
 	xorq	%r15,%r15
-	movq	-96(%r12),%xmm0
-	movq	-32(%r12),%xmm1
-	pand	%xmm4,%xmm0
-	movq	32(%r12),%xmm2
-	pand	%xmm5,%xmm1
 
 	movq	(%rsp),%r10
 	movq	%r8,%rbp
@@ -485,18 +758,9 @@ bn_mul4x_mont_gather5:
 	movq	(%rcx),%rax
 	adcq	$0,%rdx
 
-	movq	96(%r12),%xmm3
-	pand	%xmm6,%xmm2
-	por	%xmm1,%xmm0
-	pand	%xmm7,%xmm3
-
 	imulq	%r10,%rbp
 	movq	%rdx,%r11
 
-	por	%xmm2,%xmm0
-	leaq	256(%r12),%r12
-	por	%xmm3,%xmm0
-
 	mulq	%rbp
 	addq	%rax,%r10
 	movq	8(%rsi),%rax
@@ -628,7 +892,6 @@ bn_mul4x_mont_gather5:
 	movq	%r13,-24(%rsp,%r15,8)
 	movq	%rdx,%r13
 
-.byte	102,72,15,126,195
 	movq	%rdi,-16(%rsp,%r15,8)
 
 	xorq	%rdi,%rdi
@@ -712,6 +975,7 @@ bn_mul4x_mont_gather5:
 	movdqu	%xmm2,16(%rdi,%r14,1)
 	movq	8(%rsp,%r9,8),%rsi
 	movq	$1,%rax
+
 	movq	(%rsi),%r15
 	movq	8(%rsi),%r14
 	movq	16(%rsi),%r13
@@ -744,42 +1008,167 @@ bn_scatter5:
 .type	bn_gather5,@function
 .align	16
 bn_gather5:
-	movq	%rcx,%r11
-	shrq	$3,%rcx
-	andq	$7,%r11
-	notq	%rcx
-	leaq	.Lmagic_masks(%rip),%rax
-	andq	$3,%rcx
-	leaq	96(%rdx,%r11,8),%rdx
-	movq	0(%rax,%rcx,8),%xmm4
-	movq	8(%rax,%rcx,8),%xmm5
-	movq	16(%rax,%rcx,8),%xmm6
-	movq	24(%rax,%rcx,8),%xmm7
+.LSEH_begin_bn_gather5:
+
+.byte	0x4c,0x8d,0x14,0x24			
+.byte	0x48,0x81,0xec,0x08,0x01,0x00,0x00	
+	leaq	.Linc(%rip),%rax
+	andq	$-16,%rsp
+
+	movd	%ecx,%xmm5
+	movdqa	0(%rax),%xmm0
+	movdqa	16(%rax),%xmm1
+	leaq	128(%rdx),%r11
+	leaq	128(%rsp),%rax
+
+	pshufd	$0,%xmm5,%xmm5
+	movdqa	%xmm1,%xmm4
+	movdqa	%xmm1,%xmm2
+	paddd	%xmm0,%xmm1
+	pcmpeqd	%xmm5,%xmm0
+	movdqa	%xmm4,%xmm3
+
+	paddd	%xmm1,%xmm2
+	pcmpeqd	%xmm5,%xmm1
+	movdqa	%xmm0,-128(%rax)
+	movdqa	%xmm4,%xmm0
+
+	paddd	%xmm2,%xmm3
+	pcmpeqd	%xmm5,%xmm2
+	movdqa	%xmm1,-112(%rax)
+	movdqa	%xmm4,%xmm1
+
+	paddd	%xmm3,%xmm0
+	pcmpeqd	%xmm5,%xmm3
+	movdqa	%xmm2,-96(%rax)
+	movdqa	%xmm4,%xmm2
+	paddd	%xmm0,%xmm1
+	pcmpeqd	%xmm5,%xmm0
+	movdqa	%xmm3,-80(%rax)
+	movdqa	%xmm4,%xmm3
+
+	paddd	%xmm1,%xmm2
+	pcmpeqd	%xmm5,%xmm1
+	movdqa	%xmm0,-64(%rax)
+	movdqa	%xmm4,%xmm0
+
+	paddd	%xmm2,%xmm3
+	pcmpeqd	%xmm5,%xmm2
+	movdqa	%xmm1,-48(%rax)
+	movdqa	%xmm4,%xmm1
+
+	paddd	%xmm3,%xmm0
+	pcmpeqd	%xmm5,%xmm3
+	movdqa	%xmm2,-32(%rax)
+	movdqa	%xmm4,%xmm2
+	paddd	%xmm0,%xmm1
+	pcmpeqd	%xmm5,%xmm0
+	movdqa	%xmm3,-16(%rax)
+	movdqa	%xmm4,%xmm3
+
+	paddd	%xmm1,%xmm2
+	pcmpeqd	%xmm5,%xmm1
+	movdqa	%xmm0,0(%rax)
+	movdqa	%xmm4,%xmm0
+
+	paddd	%xmm2,%xmm3
+	pcmpeqd	%xmm5,%xmm2
+	movdqa	%xmm1,16(%rax)
+	movdqa	%xmm4,%xmm1
+
+	paddd	%xmm3,%xmm0
+	pcmpeqd	%xmm5,%xmm3
+	movdqa	%xmm2,32(%rax)
+	movdqa	%xmm4,%xmm2
+	paddd	%xmm0,%xmm1
+	pcmpeqd	%xmm5,%xmm0
+	movdqa	%xmm3,48(%rax)
+	movdqa	%xmm4,%xmm3
+
+	paddd	%xmm1,%xmm2
+	pcmpeqd	%xmm5,%xmm1
+	movdqa	%xmm0,64(%rax)
+	movdqa	%xmm4,%xmm0
+
+	paddd	%xmm2,%xmm3
+	pcmpeqd	%xmm5,%xmm2
+	movdqa	%xmm1,80(%rax)
+	movdqa	%xmm4,%xmm1
+
+	paddd	%xmm3,%xmm0
+	pcmpeqd	%xmm5,%xmm3
+	movdqa	%xmm2,96(%rax)
+	movdqa	%xmm4,%xmm2
+	movdqa	%xmm3,112(%rax)
 	jmp	.Lgather
-.align	16
-.Lgather:
-	movq	-96(%rdx),%xmm0
-	movq	-32(%rdx),%xmm1
-	pand	%xmm4,%xmm0
-	movq	32(%rdx),%xmm2
-	pand	%xmm5,%xmm1
-	movq	96(%rdx),%xmm3
-	pand	%xmm6,%xmm2
-	por	%xmm1,%xmm0
-	pand	%xmm7,%xmm3
-	por	%xmm2,%xmm0
-	leaq	256(%rdx),%rdx
-	por	%xmm3,%xmm0
 
+.align	32
+.Lgather:
+	pxor	%xmm4,%xmm4
+	pxor	%xmm5,%xmm5
+	movdqa	-128(%r11),%xmm0
+	movdqa	-112(%r11),%xmm1
+	movdqa	-96(%r11),%xmm2
+	pand	-128(%rax),%xmm0
+	movdqa	-80(%r11),%xmm3
+	pand	-112(%rax),%xmm1
+	por	%xmm0,%xmm4
+	pand	-96(%rax),%xmm2
+	por	%xmm1,%xmm5
+	pand	-80(%rax),%xmm3
+	por	%xmm2,%xmm4
+	por	%xmm3,%xmm5
+	movdqa	-64(%r11),%xmm0
+	movdqa	-48(%r11),%xmm1
+	movdqa	-32(%r11),%xmm2
+	pand	-64(%rax),%xmm0
+	movdqa	-16(%r11),%xmm3
+	pand	-48(%rax),%xmm1
+	por	%xmm0,%xmm4
+	pand	-32(%rax),%xmm2
+	por	%xmm1,%xmm5
+	pand	-16(%rax),%xmm3
+	por	%xmm2,%xmm4
+	por	%xmm3,%xmm5
+	movdqa	0(%r11),%xmm0
+	movdqa	16(%r11),%xmm1
+	movdqa	32(%r11),%xmm2
+	pand	0(%rax),%xmm0
+	movdqa	48(%r11),%xmm3
+	pand	16(%rax),%xmm1
+	por	%xmm0,%xmm4
+	pand	32(%rax),%xmm2
+	por	%xmm1,%xmm5
+	pand	48(%rax),%xmm3
+	por	%xmm2,%xmm4
+	por	%xmm3,%xmm5
+	movdqa	64(%r11),%xmm0
+	movdqa	80(%r11),%xmm1
+	movdqa	96(%r11),%xmm2
+	pand	64(%rax),%xmm0
+	movdqa	112(%r11),%xmm3
+	pand	80(%rax),%xmm1
+	por	%xmm0,%xmm4
+	pand	96(%rax),%xmm2
+	por	%xmm1,%xmm5
+	pand	112(%rax),%xmm3
+	por	%xmm2,%xmm4
+	por	%xmm3,%xmm5
+	por	%xmm5,%xmm4
+	leaq	256(%r11),%r11
+	pshufd	$78,%xmm4,%xmm0
+	por	%xmm4,%xmm0
 	movq	%xmm0,(%rdi)
 	leaq	8(%rdi),%rdi
 	subq	$1,%rsi
 	jnz	.Lgather
+
+	leaq	(%r10),%rsp
 	.byte	0xf3,0xc3
 .LSEH_end_bn_gather5:
 .size	bn_gather5,.-bn_gather5
 .align	64
-.Lmagic_masks:
-.long	0,0, 0,0, 0,0, -1,-1
-.long	0,0, 0,0, 0,0,  0,0
+.Linc:
+.long	0,0, 1,1
+.long	2,2, 2,2
 .byte	77,111,110,116,103,111,109,101,114,121,32,77,117,108,116,105,112,108,105,99,97,116,105,111,110,32,119,105,116,104,32,115,99,97,116,116,101,114,47,103,97,116,104,101,114,32,102,111,114,32,120,56,54,95,54,52,44,32,67,82,89,80,84,79,71,65,77,83,32,98,121,32,60,97,112,112,114,111,64,111,112,101,110,115,115,108,46,111,114,103,62,0
diff --git a/sys/conf/newvers.sh b/sys/conf/newvers.sh
index eb1b8c575873..c3cdf52a0cda 100644
--- a/sys/conf/newvers.sh
+++ b/sys/conf/newvers.sh
@@ -32,7 +32,7 @@
 
 TYPE="FreeBSD"
 REVISION="10.2"
-BRANCH="RELEASE-p12"
+BRANCH="RELEASE-p13"
 if [ "X${BRANCH_OVERRIDE}" != "X" ]; then
 	BRANCH=${BRANCH_OVERRIDE}
 fi